What's Guilt (or Deterrence) Got to Do With It?: The Death

UIC School of Law UIC School of Law

UIC Law Open Access Repository UIC Law Open Access Repository

UIC Law Open Access Faculty Scholarship

1-1-1997

What's Guilt (or Deterrence) Got to Do With It?: The Death Penalty, What's Guilt (or Deterrence) Got to Do With It?: The Death Penalty,

Ritual, and Mimetic Violence, 38 Wm. & Mary L. Rev. 487 (1997) Ritual, and Mimetic Violence, 38 Wm. & Mary L. Rev. 487 (1997)

Donald L. Beschle The John Marshall Law School, [email protected]

Follow this and additional works at: https://repository.law.uic.edu/facpubs

Part of the Criminal Law Commons, Criminal Procedure Commons, Law Enforcement and Corrections

Commons, and the Public Law and Legal Theory Commons

Recommended Citation Recommended Citation Donald L. Beschle, What's Guilt (or Deterrence) Got to Do With It?: The Death Penalty, Ritual, and Mimetic Violence, 38 Wm. & Mary L. Rev. 487 (1997).

https://repository.law.uic.edu/facpubs/188

This Article is brought to you for free and open access by UIC Law Open Access Repository. It has been accepted for inclusion in UIC Law Open Access Faculty Scholarship by an authorized administrator of UIC Law Open Access Repository. For more information, please contact [email protected].

https://repository.law.uic.edu/

https://repository.law.uic.edu/facpubs

https://repository.law.uic.edu/facpubs?utm_source=repository.law.uic.edu%2Ffacpubs%2F188&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/912?utm_source=repository.law.uic.edu%2Ffacpubs%2F188&utm_medium=PDF&utm_campaign=PDFCoverPages





mailto:[email protected]

WHATS GUILT (OR DETERRENCE) GOT TO DO WITH IT?:THE DEATH PENALTY, RITUAL, AND MIMETIC VIOLENCE

DONALD L. BESCHLE"

INTRODUCTION

For decades, the death penalty has been one of the most pas-sionately debated topics in American law. Lawyers, social scien-tists, philosophers, theologians, the mass media, and ordinarycitizens have argued over its wisdom and legitimacy. Remark-ably, though, when the principal arguments for and against thedeath penalty are examined closely, they seem inadequate to thetask of either justifying the death penalty or proving convincing-ly that it must be abolished.

As a question of constitutional law, the death penalty debateunsurprisingly has focused on whether executions are "cruel andunusual" within the meaning of the Eighth Amendment.' Beyondquestions of how to give specific content to those general terms,the government's choice of criminal punishments must meet amore fundamental requirement. At the very least, due process oflaw requires that there be a rational basis for government actionthat deprives one of life or liberty.2 Thus, rational argumentmust support the choice of punishment. Of course, the choice of

* Associate Professor, The John Marshall Law School; B.A. Fordham University;J.D. New York University School of Law; LL.M. Temple University School of Law.

1. "Excessive bail shall not be required, nor excessive fines imposed, nor crueland unusual punishments inflicted." U.S. CONST. amend. VIII.

2. Although Justices have disagreed sharply over how stringent the burden is up-on government to demonstrate that its actions are rational, even the most deferen-tial would maintain that some minimal rationality must be established. For example,in Lochner v. New York, 198 U.S. 45 (1905), the dissenters did not challenge thenotion that the state of New York must act reasonably in its economic regulation;rather, they applied a more permissive standard of determining reasonableness. Seeid. at 76 (Holmes, J., dissenting); id. at 66-74 (Harlan, J., with whom White & Day,JJ., join, dissenting). Professor Cass Sunstein maintains: "Above all, the AmericanConstitution was designed to create a deliberative democracy ... The minimal con-dition of deliberative democracy is a requirement of reasons for government action."CAsS H SUNSTEIN, THE PARTIAL CONSTITUTION 19-20 (1993).

487

HeinOnline -- 38 Wm. & Mary L. Rev. 487 1996-1997

WILLIAM AND MARY LAW REVIEW

punishment never has been subject to strict scrutiny, whichwould impose a requirement that the punishment go no furtherthan necessary to achieve the government's interest.' Still, thechoice by government to impose a severe punishment may not bearbitrary; rational grounds for the choice are required.

The search for a rational basis for the death penalty has ledto two primary, alternate justifications, both of which abolition-ists contest.4 The first of these is the consequentialist argumentthat imposition of the death penalty furthers clearly defined pur-poses, usually deterrence, better than do less severe punish-ments.5 Abolitionists seem most comfortable contesting this lineof argument." Although proponents of the death penalty by nomeans concede an inability to demonstrate the validity of theirconsequentialist arguments,7 the weight of evidence seems torefute proponents' claims of deterrence.'

When faced with the weaknesses of deterrence-basedconsequentialist arguments, death penalty proponents gravitatetoward arguments based on retributive theory.' According tothis line of thought, desert and individual responsibility justify,if not demand, the symmetry of capital punishment as a re-sponse to murder, irrespective of the existence or nonexistence ofany deterrence."0 Though some abolitionists have tried to frametheir case upon retributivist thought, this line of argument is

3. Such a requirement would, more or less, constitutionalize Jeremy Bentham'sclassic utilitarian conclusions regarding punishment. Because the end of law "is toaugment. . . happiness" and "punishment... is [an] evil," punishment should notbe imposed in excess of its ability "to exclude some greater evil." JEREMY BENTHAM,AN INTRODUCTION TO THE PRINCIPLES OF MORALS AND LEGISLATION 170 (WilfiidHarrison ed., Basil Blackwell & Mott, Ltd. 1948) (1789).

4. See, e.g., WALTER BERNS, FOR CAPITAL PUNISHMENT' CRIME AND THE MORAIxTYOF THE DEATH PENALTY 83-152 (1991); see generally MARVIN HENBERG, RETRIUTION:EVIL FOR EVIL IN ETHICS, LAW, AND LITERATURE (1990) (exploring retributivethought).

5. See BERNS, supra note 4.6. See MARK TUSHNET, THE DEATH PENALTY 5-7 (1994).7. See id.8. See, e.g., KILMAN SHIN, DEATH PENALTY AND CRIME 1-71 (1978); FRANKLIN E.

ZIMRING & GORDON J. HAWKINS, DETERRENCE: THE LEGAL THREAT IN CRIME CON-TROL 186-90 (1973).

9. See HENBERG, supra note 4.10. See id. at 221-28.

488 [Vol. 38:487


DEATH PENALTY AND MIMETIC VIOLENCE

used primarily by death penalty advocates."Recent Supreme Court cases seem to take the position that

the state need not win the consequentialist debate in order todefend the death penalty. 2 Retribution, it would appear, pro-vides a sufficient foundation. Analysis of the Court's death pen-alty jurisprudence, however, shows a surprising lack of sympa-thy for arguments that seem to go directly to the core concernsof retributivism, the assessment of personal responsibility, andeven the question of guilt itself." Perhaps these cases indicatethat something else is going on here, something of which theJustices themselves may be largely unaware, something neitherretributivist nor consequentialist, and perhaps somewhat beyondthe limits of traditional rational argument.

Rend Girard, a theorist whose work attempts to connect liter-ary criticism, anthropology, and theology, contends that religion,law, and indeed many of the bonds of civilization and culture,rely on ritualized violence in order to break and tame the cycleof imitative, or mimetic, violence that inevitably arises withinsociety. 4 Girard's theories evoke controversy, but if they areeven partially correct, they may shed a great deal of light on therole played by the death penalty in American society. They mayexplain much that recourse to either consequentialism orretributivist theory leaves unexplained.

If Girard's theories do help explain the persistence of thedeath penalty, they also may require those who debate the issueto confront new questions. Defenders of the death penalty, onthe one hand, will be hard pressed to establish that, at least asit is currently administered, it is not weighted down with as-pects that are not merely inadequate to satisfy aconsequentialist, but perhaps are beyond the scope of what wecan accept as rational, even under retributivist theory. Aboli-tionists, on the other hand, will be challenged by the apparentneed of society for ritualized community violence. In otherwords, is abolition possible without the creation of some alterna-

11. See TUSHNET, supra note 6, at 4-7.12. See generally id. at 113-17 (discussing recent death penalty decisions by the Court).13. See id.14. See REN9 GIRARD, VIOLENCE AND THE SACRED (Patrick Gregory trans., Johns

Hopkins University Press 1977) (1972).

1997]



tives to satisfy a powerful social need, regardless of whether ra-tional or nonrational foundations form the bases for that need?

Part I of this Article gives a necessarily brief overview of Su-preme Court death penalty jurisprudence over the last decades,focusing on the main lines of consequentialist and retributivistargument put forward by abolitionists and death penalty propo-nents. Part II explores the ultimate inability of eitherconsequentialist or retributivist arguments to explain significantparts of current death penalty jurisprudence satisfactorily. PartIII sketches Girard's theories of mimetic violence and the role ofthe scapegoat, and applies these ideas to explain the role of thedeath penalty in American society. Finally, Part IV discusses theimplications of accepting an explanatory model based uponGirard's thought for the future of the death penalty debate.

I. THE DEATH PENALTY AND THE SUPREME COURT:A BRIEF HISTORY

The Eighth Amendment provides that the federal governmentmay not impose "cruel and unusual punishments."15 Morebroadly, the Fifth Amendment provides that no person may be"deprived of life, liberty, or property without due process oflaw."" The Fourteenth Amendment extends the due processguarantee to the states, and the Supreme Court has held thatthe Eighth Amendment is incorporated in that guarantee, andthus also binds the states. 7

Before the 1960s, the Supreme Court considered a number ofissues regarding the procedure under which the death penaltywas imposed, but did not seriously entertain the argument thatthe penalty was itself unconstitutional."8 In 1889, the Court re-jected the argument that the new electric chair was a cruel and

15. U.S. CONST. amend. VIII.16. Id. amend. V.17. See Robinson v. California, 370 U.S. 660 (1962) (incorporating the proscription of

cruel and unusual punishment through the Fourteenth Amendment Due Process Clause).18. See Francis v. Resweber, 329 U.S. 459 (1947) (rejecting a due process and

Eighth Amendment challenge to the ability of a state to attempt to execute a pris-oner after the first attempt had failed); In re Kemmler, 136 U.S. 436 (1890) (reject-ing a due process challenge).

[Vol. 38:487490



unusual means of imposing the death penalty,19 and in a grisly1947 case, a sharply divided Court held that it was not crueland unusual to electrocute a murderer a second time when thechair malfunctioned and delivered an insufficient shock the firsttime."0 In 1931, however, the Court held that a defendant fac-ing the death penalty was entitled, as other criminal defendantsat the time were not, to appointed counsel.2

In 1963, Justice Arthur Goldberg circulated a memorandum tohis colleagues urging that the Court select several cases for re-view as vehicles to determine "[w]hether, and under what cir-cumstance, the imposition of the death penalty is proscribed" bythe Constitution.22 In light of the abolition of the penalty bymost western democracies and declining public support at thetime within the United States, Justice Goldberg contended thatthe issue was ripe for consideration.' He also stated that in hisopinion, the death penalty had to be justified as an effective de-terrent, because vengeance was not "an acceptable goal ofpunishment."24 Justice Goldberg's memorandum did not per-suade the Court to grant review, but it generally is regarded asthe starting point of the contemporary history of the Court'sstruggle with this issue.

Between 1963 and 1972, the Supreme Court considered sever-al death penalty cases, rejecting challenges based both upon the,absence of clear standards to guide juries in choosing the penal-ty' and upon the need for bifurcated proceedings to address

19. See In re Kemmler, 136 U.S. at 447.20. See Francis, 329 U.S. at 463. For a full account of the case, see ARTHUR S.

MILLER & JEFFREY H. BOWfAN, DEATH BY INSTALLMENTS: THE ORDEAL OF WALMFRANCIS (1988).

21. See Powell v. Alabama, 287 U.S. 45 (1932). This case was the famous case of the"Scottsboro Boys," nine young blacks accused of raping two white women. The casecommanded national attention. See generally JAMES GOODMAN, STORIES OF SCOTTSBORO(1994) (describing the trial of the "Scottsboro Boys" and the public reaction).

22. Memorandum from Arthur J. Goldberg to the Conference Re: Capital Punish-ment, October Term, 1963, reprinted in TUSHNET, supra note 6, at 149-62.

23. See TUSHNET, supra note 6, at 155-58.24. Id. at 159. Justice Goldberg contended that the burden of proving the existence

of a deterrent effect was on the state, and that, at best, the evidence was inconclusive.See id. at 159-60. The state's burden, therefore, had not been met. See id.

25. See, for example, McC'autha v. California, 402 U.S. 183 (1971), which heldthat it was not unconstitutional to "commit[ I to the untrammeled discretion of the

491199'7]


492 WILLIAM AND MARY LAW REVIEW [Vol. 38:487

guilt and punishment.26 Several Justices dissented, however,and, for the first time, significant abolitionist sentiment ap-peared in the pages of the United States Reports. During thisperiod, state courts in California"5 and New Jersey' ruledthat the death penalty violated state constitutional provisions.

Finally, in 1972, the Court invalidated the death penalty asapplied under then-existing statutes.30 Furman v. Georgia wasdecided not only by the narrowest of margins, but also featuredfive separate opinions from the five Justices in the majority.3'Justices Brennan and Marshall found the death penalty to beunconstitutional per se;32 Justices Douglas, Stewart, and Whitefound that its application was arbitrary and discriminatory, butdid not go so far as to state that death penalties could not beframed in a way that would satisfy constitutional standards."3

The connecting thread running through the five opinions wasthe principle that punishment must be rational.' Although

jury the power to pronounce life or death in capital cases." Id. at 207. InWitherspoon u. Illinois, 391 U.S. 510 (1968), however, the Court held that a statecould not enforce a blanket exclusion of jurors with anti-capital punishment senti-ments from juries in capital cases, but in doing so, recognized and implicitly en-dorsed the fact that juries might be given wide discretion in imposing the deathpenalty. See id. at 518-21.

26. See MeGautha, 402 U.S. at 208-20. The argument in favor of bifurcation isthat a defendant cannot be placed in a position of having to waive his right to besilent on the issue of his guilt in order to introduce evidence that would serve tomitigate the severity of his punishment. See id. at 210-11.

27. See id. at 226-312 (Douglas, J., with whom Brennan & Marshall, JJ., join,dissenting).

28. See People v. Anderson, 493 P.2d 880 (Cal. 1972).29. See State v. Funicello, 286 A.2d 55 (N.J. 1972).30. See Furman v. Georgia, 408 U.S. 238 (1972) (per curiam).31. See id.32. "When examined by the principles applicable under the Cruel and Unusual

Punishments Clause, death stands condemned as fatally offensive to human dignity.The punishment of death is therefore 'cruel and unusual,' and the States may nolonger inflict it as a punishment for crimes." Id. at 305 (Brennan, J., concurring)."[Tlhe death penalty is an excessive and unnecessary punishment that violates theEighth Amendment." Id. at 358-59 (Marshall, J., concurring).

33. Justice Douglas was concerned with selective application of the death penalty,particularly its impact on minority defendants. See id. at 249-57 (Douglas, J., concur-ring). Justice Stewart was concerned with the arbitrary way in which the penaltywas imposed, see id. at 306-10 (Stewart, J., concurring), as was Justice White, seeid. at 310-14 (White, J., concurring).

34. See id. at 249 (Douglas, J., concurring); id. at 274 (Brennan, J., concurring);


19971 DEATH PENALTY AND MIMETIC VIOLENCE

death penalty litigation often is thought to revolve around theEighth Amendment, the Due Process Clauses of the Fifth andFourteenth Amendments also imply this primary requirement ofrationality.35 The basic test of the rationality of a governmentact is determining whether it relates to achieving a legitimategovernment end.36 Inquiry must begin, then, with an inventoryof the legitimate ends of punishment. The most commonly notedpurposes of punishment are incapacitation,37 deterrence ofothers,38 rehabilitation of the offender,39 and retribution-themost difficult to define clearly.40 Certainly, a punishment thataccomplished none of these goals would be vulnerable to attackas irrational. Because criminal punishment is an explicit con-cern of the Eighth Amendment, one would be justified in de-manding, if not the extremely close fit between means and endsthat strict scrutiny demands, then at least something more thana minimum suggestion of rationality.41

id. at 310 (Stewart, J., concurring); id. at 312-13 (White, J., concurring; id. at 331(Marshall, J., concurring).

35. See SUNSTEIN, supra note 2. Professor Tribe has noted that even in abandon-ing the strict scrutiny given to means-end relationships during the era of Lochner v.New York, 198 U.S. 45 (1905), "the Court never wholly abandoned the position thatlegislatures ... must always act in furtherance of public goals." LAURENCE H.TRIBE, AMERICAN CONSTITUTIONAL LAw 582 (2d ed. 1988).

36. See TRIBE, supra note 35, at 582-83.37. See NIGEL WALKER, WHY PUNISH? 34-41 (1991). Incapacitation, that is, render-

ing the individual offender unable to repeat his crime, is usually a temporary mat-ter, running the length of the offender's prison term. See id. at 34. Thus, "most sen-tences merely postpone opportunities for reoffending." Id. Of course, the longer aprison term, the closer a sentence gets to substantially, if not entirely, incapacitatinga prisoner.

38. See generally ZIMERING & HAWKINS, supra note 8. Deterrence usually is dis-cussed in terms of its effect on others, rather than on the offender, although theoffender also may be deterred in the future. See id. at 18-33.

39. See PHILIP BEAN, PUNISHiMENT 53-65 (1981). Although rehabilitation generallyis seen as the most modem theory of punishment, it can trace its philosophical rootsto Plato and St. Thomas Aquinas, among others, who saw "[w]ickedness [as] a men-tal disease," and "punishment ... as a moral medicine." Id. at 54.

40. See the excerpts from Kant, Hegel, and others collected in PHILOSOPHICALPERSPECTIVES ON PUNISHMENT 102-35 (Gertrude Ezorsky ed., 1977) [hereinafterPHILOSOPHICAL PERSPECTIVES]; see also the extensive treatment of the subject inHENBERG, supra note 4.

41. The requisite closeness of the fit between legislative means and ends is one ofthe most pervasive issues in constitutional law. See SUNSTEIN, supra note 2; TRIBE,supra note 35.



To Justice Brennan, punishment became arbitrary, and thusunconstitutional, when it exceeded the severity necessary toachieve any of the accepted goals of punishment.42 The deathpenalty obviously does not seek to rehabilitate. Justice Brennanstated that life imprisonment could achieve incapacitation, andthat the "remote and improbable" risk of death was unlikely tohave a significant deterrent effect.43 As for retribution, whichJustice Brennan described as "[tihe asserted public belief thatmurderers... deserve to die,"" the fact that the overwhelmingmajority of murderers were not punished by death led him tothe conclusion that "the execution of a random few" did not fur-ther the retribution goal significantly.45

Justice Marshall's opinion is probably best known for its rejec-tion of the notion that uninformed public opinion could suffice toestablish the constitutional acceptability of the death penalty.46

For the purposes of this Article, however, it is more significantthat Justice Marshall alone among the Justices explicitly reject-ed retribution's legitimacy as a goal of punishment."' Overstat-ing the empirical case, he maintained that "no one has ever seri-ously advanced retribution as a legitimate goal" in contemporaryAmerica, but that the debate concerning capital punishment was"always mounted on deterrent" arguments. 48 Perhaps more in-

42.A punishment is excessive . . . if it is unnecessary: The infliction of asevere punishment by the State cannot comport with human dignitywhen it is nothing more than the pointless infliction of suffering. If thereis a significantly less severe punishment adequate to achieve the purpos-es for which the punishment is inflicted . . . the punishment inflicted isunnecessary and therefore excessive.

Furman v. Georgia, 408 U.S. 238, 279 (1972) (Brennan, J., concurring). Here, JusticeBrennan essentially puts forward the position of Jeremy Bentham: Because "all pun-ishment in itself is evil . . . it ought only to be admitted -in as far as it promises toexclude some greater evil," and "[it] ought not to be inflicted .... where the mis-chief may be prevented ...without it: that is, at a cheaper rate." BENTHAM, supranote 3, at 158-62.

43. Furman, 408 U.S. at 302 (Brennan, J., concurring).44. Id. at 304 (Brennan, J., concurring).45. Id. at 304-05 (Brennan, J., concurring).46. See id. at 329-33 (Marshall, J., concurring).47. See id. at 342-45 (Marshall, J., concurring).48. Id. at 363 (Marshall, J., concurring). Although consequentialist arguments were

more prominent in the years preceding Furman, retribution was not entirely without

[Vol. 38:487494



teresting was his assertion that to accept retribution as a legiti-mate goal would render the Eighth Amendment essentially un-enforceable: "all penalties selected by the legislature would bydefinition be acceptable means for designating society's moralapprobation of a particular act."49

Justices Douglas, White, and Stewart did not reject the notionthat the death penalty might be defended on rational grounds,but they found fatal flaws in the then-current practice. The se-lective imposition of death, and the likelihood of its applicationin unacceptable discriminatory ways troubled Justice Douglas.0

Justices White and Stewart each accepted the legitimacy of ret-ribution, but found that imposition of capital punishment on a"capriciously selected random handful"5' of offenders could notmeaningfully satisfy what Justice Powell termed humankind's"retributive instincts."52

Although the four dissenters obviously found capital punish-ment to be rational and constitutionally justifiable, their opin-ions primarily sounded a theme of judicial deference to legisla-tures rather than mounting a direct defense of the death penal-ty." In fact, several Justices felt compelled to disparage the

its defenders. See, e.g., H.J. McCloskey, A Non-Utilitarian Approach to Punishment, 8INQUIRY 239 (1965), reprinted in PHILOSOPHIcAL PERSPECTiVES, supra note 40, at119-34.

49. Furman, 408 U.S. at 344 (Marshall, J., concurring).50. See id. at 250-57 (Douglas, J., concurring).51. Id. at 309-10 (Stewart, J., concurring). "[Wihen imposition of the penalty

reaches a certain degree of infrequency, it would be very doubtful that any existinggeneral need for retribution would be measurably satisfied." Id. at 311 (White, J.,concurring).

52. Id. at 454 (Powell, J., dissenting).53. "If today's opinions demonstrate nothing else, they starkly show that this is

an area where legislatures can act far more effectively than courts." Id. at 403 (Bur-ger, C.J., dissenting). "[The abolitionist argument] makes sense only in a legislativeand executive way and not as a judicial expedient." Id. at 410 (Blackmun, J., dis-senting). "The sobering disadvantage of constitutional adjudication of this magnitudeis the universality and permanence of the judgment. The enduring merit of legisla-tive action is its responsiveness to the democratic process, and to revision andchange ... ." Id. at 462 (Powell, J., dissenting).

The most expansive reading of the leading constitutional cases does notremotely suggest that this Court has been granted a roving commission,either by the Founding Fathers or by the framers of the FourteenthAmendment, to strike down laws that are based upon notions of policy ormorality suddenly found unacceptable by a majority of this Court.

1997] 495



practice: Justice Blackmun stated that as a legislator, he wouldvote against the death penalty; 4 Chief Justice Burger statedthat he might vote similarly under the same circumstances;'and Justice Powell, in the course of analyzing the goals of pun-ishment, stated that retribution was a permissible, althoughperhaps "unworthy" rationale.56

Although the abolitionist community widely regarded Furmanas the final victory, 7 the case contained the seeds of its owndemise; one that would come swiftly. Because at least two Jus-tices in the Furman majority believed that the death penaltycould be made rational, and therefore constitutionally accept-able, by being made more determinate, legislators explored waysof doing just that. The most determinate option, of course, was amandatory death penalty, but, in 1976, the Court found suchstatutes unconstitutional, at least as applied to broad categoriesof murder.58 A constitutionally acceptable death penalty, it washeld, must permit "particularized consideration" of the crimeand the defendant.59 In Gregg v. Georgia,'o however, the Courtapproved a statute that gave "guidance" and "direction" to a juryin deciding which defendants should be sentenced to death.6Once again, most Justices either expressly or implicitly endorsedboth deterrence and retribution as legitimate goals of punish-ment,62 but now it was held essential to consider the particularblameworthiness of the defendant.'

Thus, the Court has taken the position that a rational system

Id. at 467 (Rehnquist, J., dissenting).54. See id. at 406 (Blackmun, J., dissenting).55. See id. at 375 (Burger, C.J., dissenting).56. See id. at 453 (Powell, J., dissenting).57. "With Furman, the . . . campaign against the death penalty appeared to have

reached a successful . . . conclusion . . . . The Court, it might have seemed, hadabolished the death penalty." TUSHNET, supra note 6, at 54. "To many observers,Furman had, for all practical purposes, ended the practice of executions in America."FRANKiN E. ZMING & GORDON HAWKINS, CAPrrAL PUNISHMENT AND THE AMERCANAGENDA 64 (1986).

58. See Woodson v. North Carolina, 428 U.S. 280 (1976).59. Id. at 303.60. 428 U.S. 153 (1976).61. Id. at 192-98.62. See id. at 183-87; i& at 207-26 (White, J., concurring in the judgment).63. See id. at 183-87.

496 [Vol. 38:487



of retribution (and apparently also of deterrence) may not insistthat the punishment automatically mirror the criminal's act.Retribution must indicate society's response not merely to theconsequences of the act, but also to the extent to which the actorwas unusually dangerous or blameworthy. Three Justices reject-ed this view, and voted to uphold mandatory death sentences;"Justices Brennan and Marshall continued to insist that all deathpenalties were unconstitutional;' and Justice Marshall re-newed his argument that retribution was not a legitimate basisfor punishment.6

Although most of the Justices rejected the contention that ret-ribution was an illegitimate purpose of punishment, case law ex-isted to support the contention that retribution alone might besuspect.67 Deterrence appeared to be the more secure justifica-tion. Post-Gregg developments, however, suggest a distinct re-versal of that position. In 1977, the Court held that the deathpenalty as punishment for rape was unconstitutionally exces-sive.' The Court also held that the death penalty could be usedonly in cases of intentional murder; it was impermissible if ap-plied to a defendant who neither killed nor demonstrated anyintent that anyone be killed. 9 Most commentators interpretthese cases to indicate that death constitutionally may be im-posed only for murder.

64. Woodson, 428 U.S. at 306-07 (White, J., with whom Burger, C.J. & Rehnquist, J.,join, dissenting).

65. See Gregg, 428 U.S. at 227-31 (Brennan, J., dissenting); id. at 231-41 (Mar-shall, J., dissenting).

66. See id. at 23941 (Marshall, J., dissenting).67. "In part, capital punishment is an expression of society's moral outrage at par-

ticularly offensive conduct .... 'Retribution is no longer the dominant objective ofthe criminal law,' but neither is it a forbidden objective .... " Id. at 183 (citingWilliams v. New York, 337 U.S. 241, 248 (1949)).

68. See Coker v. Georgia, 433 U.S. 584, 592, 597-98 (1977) (stating that although"[s]hort of homicide, [rape] is the 'ultimate violation of self", the death penalty was"an excessive penalty for the rapist who ... does not take human life").

69. Enmund v. Florida, 458 U.S. 782, 788 (1982). Nevertheless, the Court has sub-sequently held that a participant in a felony who did not himself kill another mightreceive the death penalty under felony murder principles if he was a "major" partici-pant and displayed "reckless indifference to human life." Tison v. Arizona, 481 U.S.137, 158 (1987).

70. This seems to be the clear implication from Coker. See Coker, 433 U.S. at 597-98 (noting that "life ... is not over" for the rape victim). As Professor Mark

1997] 497



In terms of pure deterrence, of course, the conclusion that thedeath penalty can only apply in murder cases could not be sus-tained. Although the efficacy of using the death penalty to determurder may be questioned,7 no reason exists to believe that itwould deter other crimes any less. The conclusion that rape andother felonies may not be punished by death seems to indicateclearly that, perhaps surprisingly, deterrence may not be thesole rationale behind the imposition of severe punishment. Farfrom being an illegitimate justification, retribution seems to bean essential element of the modern justification of the deathpenalty.

Since 1977, the basic principles established in Gregg and thecases immediately following it have been maintained. The argu-ments that have broken out generally have accepted their basicparadigms. The death penalty itself is now impervious to consti-tutional challenge; Justices Marshall and Brennan, the consis-tent dissenters on that point,72 no longer sit on the Court.Death still requires justification as being rational, however, andwhen an individual case or a general practice suggests thatcapital punishment might be arbitrary, the Court at least hastaken challenges seriously, more so than any fundamental at-tack on the per se use of the death penalty.7"

These challenges, though not entirely ignorant of matters ofdeterrence, 4 focus primarily on the extent of the defendant's

Tushnet wrote: "Since 1977, it has been generally accepted that capital punishmentis reserved for people who commit murder; if death is an excessive punishment forrape, it appears hard to come up with a crime other than murder for which itwould not be excessive." TUSHNET, supra note 6, at 74.

71. See generally ZIvMRNG & HAWKINS, supra note 57, at 167-86 (stating that thedeterrent effect is nil or very small in relation to total homicide volume).

72. See, e.g., Gregg v. Georgia, 428 U.S. 153, 227 (1976) (Brennan, J., dissenting);id. at 231 (Marshall, J., dissenting).

73. See, e.g., Pulley v. Harris, 465 U.S. 37, 50 (1984) (determining that a Californiastatutory scheme was not "arbitrary" for failing to require proportionality review).

74. In Coker, the Court struck down the death penalty for rape "even though itmay measurably serve the legitimate ends of punishment." Coker, 433 U.S. at 593n.4. Presumably, deterrence is at least one of those ends. In Enmund, while focusingon individual culpability, the Court also noted: "We are quite unconvinced, however,that the threat that the death penalty will be imposed for murder will measurablydeter one who does not kill and has no intention or purpose that life will be taken."Enmund v. Florida, 458 U.S. 782, 798-99 (1982).

498 [Vol. 38:487


19971 DEATH PENALTY AND MIMETIC VIOLENCE 499

blameworthiness, highlighting the centrality of the retributivejustification for the death penalty. Thus, at some point theoffender's age will make death unconstitutional. The Court hasnoted that "such a young person is not capable of acting with thedegree of culpability that can justify the ultimate penalty."75

Although a mentally retarded person may be executed,7" an in-sane person may not be.7 Given the distinctions that the Courthas made, these decisions must relate more to retribution andblameworthiness than to deterrence. Unsuccessful challengeshave been framed in terms of retribution, as well. The argumentthat the death penalty is subject to racial bias7' assumes that ifcapital punishment is to exist, it should be reserved for thosemost blameworthy, untainted by extraneous factors.79

In recent years, the futility of direct attacks on the constitu-tionality of capital punishment has led to a series of proceduraldisputes focusing on the availability of habeas corpus review ofstate proceedings.'o In general, the Court has displayed grow-

75. Thompson v. Oklahoma, 487 U.S. 815, 823 (1988). The Court held that a de-fendant who was 15 years old at the time of the crime could not be executed. Seeid. Even the dissenters, although unwilling to set the age at 16, agreed that "atsome age a line does exist ... below which a juvenile can never be considered fullyresponsible for murder." Id. at 872 (Scalia, J., dissenting).

76. See Penry v. Lynaugh, 492 U.S. 302, 338-39 (1989) ("In light of the diversecapacities and life experiences of mentally retarded persons, it cannot be said on therecord before us today that all mentally retarded people, by definition, can never actwith the level of culpability associated with the death penalty.").

77. See Ford v. Wainwright, 477 U.S. 399 (1986).78. See, e.g., McCleskey v. Kemp, 481 U.S. 279 (1987) (rejecting a statistically

based argument that Georgia's death penalty was administered in a racially discrim-inatory way).

79. McCleskey argued, for example, not only that black defendants were morelikely than white defendants to be sentenced to death, but also that the race of thevictim was highly significant. See id. at 321 ("[Clases involving black defendants andwhite victims are more likely to result in a death sentence than cases featuring anyother racial combination of defendant and victim .... ") (Brennan, J., dissenting).

80. See J. Thomas Sullivan, 'Reforming" Federal Habeas Corpus: The Cost to Fed-eralism; The Burden for Defense Counsel; and the Loss of Innocence, 61 UMKC L.REV. 291 (1992) (discussing the Court's movement toward a more restrictive view ofthe writ of habeas corpus); see also Joseph L. Hoffman, Is Innocence Sufficient? AnEssay on the U.S. Supreme Court's Continuing Problems with Federal Habeas Corpusand the Death Penalty, 68 IND. L.J. 817, 818 (1993) (noting that recent. EighthAmendment jurisprudence has taken a "process oriented" approach focusing on deathpenalty procedures, rather than a "substantive" approach focusing on the extent towhich particular death sentences are deserved).



ing impatience with the length of postconviction proceedings,and has acted to expedite the actual imposition of the penalty."'Many of the particulars of these disputes are unnecessary forthis Article's purposes, but, as will be discussed below, the gen-eral tone of impatience has led to at least a few decisions thatshould trouble those who are committed to defending capitalpunishment based on retributive theory. At this point, it will beuseful to step back from the specific decisions of the SupremeCourt and discuss more broadly the rationales that have beenused to justify or attack capital punishment over the years.

II. RATIONALIZING PUNISHMENT

Evidence of the existence of a sense that some limits must beplaced on the range and severity of punishment imposed uponcriminals preceded the drafting of the Eighth Amendment. 2

Enlightenment reformers debated the question of punishment,and insisted that it must be justified as producing some recog-nizable benefit." They viewed the imposition of suffering asbad, and therefore not a legitimate end in itself." Althoughreformers acknowledged the fact that the act of imposing pun-ishment might itself produce pleasure, that sort of "benefit" wasunworthy of rational human beings.' Under this view, to pun-

81. Professor Sullivan has noted that lt]he political and judicial outcry for 'reform'of federal habeas corpus practice has been mirrored in a series of recent SupremeCourt decisions effectively streamlining the process for seeking federal relief fromstate court convictions." Sullivan, supra note 80, at 291 (citing Delo v. Stokes, 495U.S. 320 (1990); Woodard v. Hutching, 464 U.S. 377 (1984)).

82. - As early as 1809, a volume of the views of various prominent thinkers on thequestion of capital punishment could be compiled, including, among others, Beccaria,Montesquieu, Thomas More, Erasmus, Benjamin Franklin, Samuel Johnson, andJeremy Bentham. See generally BASIL MONTAGU, THE OPINIONS OF DIFFERENT AU-THORS UPON THE PUNISHMENT OF DEATH (reprint, William S. Hein & Co. 1984) (1809).

83. See id.84. See BENTHAM, supra, note 3, at 29 (outlining his formulation of the principle).85. Benjamin Franklin, arguing against the common 18th century practice of exe-

cuting thieves, analogized excessive punishment to crime itself. "To put a man todeath for a crime which does not deserve death, is it not murder?" Letter from Dr.Benjamin Franklin to Benjamin Vaughn, Esq. On the Criminal Laws, and the Prac-tice of Privateering (Mar. 14, 1785), in 1 MONTAGU, supra note 82, at 161. He wenton to describe an execution for minor theft as "barbarous . . . bloody-minded, andrevengeful." Id. at 164.

[Vol. 38:487500



ish for no other reason than to enjoy the imposition of pain wasbestial.86 The punisher, under those circumstances, became nobetter than the criminal. 7

By no means were the reformers without opposition. Somecommentators defended unbridled vengeance as necessary toaddress some fundamental human instinct.'m The nineteenthcentury, however, saw the fundamental victory of the positionthat the imposition of punishment, and its degree, had to be jus-tified. Physical punishments and banishment largely gave wayto prison sentences. 9 Special treatment for juveniles and theinsane began to take root as well." To a large extent, this post-Enlightenment trend was committed to rehabilitation as its pri-mary rationale; thus the particular characteristics of the individ-ual convict were of central importance."

Perhaps out of frustration at its limited success, rehabilitationlargely has faded from popular discussion of rationales for pun-ishment, with the possible exception of cases involving juve-niles. 2 Of course, the death penalty does not, and never has,found its justification in rehabilitation. Rehabilitation has beenrelevant to the death penalty debate only insofar as abolitionistshave argued that the obvious irrelevance of the penalty to reha-bilitative ends calls for its rejection. 3 Thus, for purposes of this

86. See id.87. See id.88. For a discussion of the pervasiveness of retributive thought, see HENBERG, su-

pra note 4, at 21-42. Henberg discusses some of the extremes to which the urge toretribution can lead, such as the historic practice of trying and executing animals forthe damage they caused, as well as the possibility that there may be a genetic basisfor the urge to retaliate. See id.

89. See LAWRENCE M. FRIEDMAN, CRIME AND PUNISHMENT IN AMERICAN HISTORY74-82 (1993).

90. See id. at 163-66 (discussing the development of separate juvenile justice sys-tems); id, at 140-48 (discussing the insanity defense and related issues).

91. For discussion of the development of prisons in the late-18th and early-19thcenturies, and the rise and fall of the rehabilitative ideal, see Randall McGowan,The Well-Ordered Prison: England, 1780-1865, in THE OXFORD HISTORY OF THE PRIS-ON 79-109 (Norval Morris & David J. Rothman eds., 1995); and David J. Rothman,Perfecting the Prison: United States 1789-1865, in THE OXFORD HISTORY OF THEPRISON, supra, at 111-29. For a discussion of the rehabilitative ideal of the "reformschool" for juvenile corrections, see Steven Schlossman, Delinquent Children: TheJuvenile Reform School, in THE OXFORD HISTORY OF THE PRISON, supra, at 363-89.

92. See BEAN, supra note 39, at 169.93. One might argue that execution "rehabilitates" the offender in that, by paying

1997] 501



Article, rehabilitation as a legitimate goal can be put aside.A second rational goal of punishment is incapacitation of the

defendant, that is, assuring that he is unable to repeat hiscrime. Surely, the death penalty incapacitates, but this justifica-tion of capital punishment does not seem prominently advancedin either scholarly or popular debate. Life imprisonment effec-tively achieves incapacitation nearly as well as does death. 4

The rare instance of a murder in prison committed by a convictalready serving a life sentence does not seem to generate greatpublic concern and a death penalty limited to this small numberof cases will not satisfy death penalty proponents. Most currentdebate over capital punishment therefore turns on two broadjustifications: deterrence of others and retribution.

The argument connected most obviously to rational thought,with its insistence upon a demonstration of the connection be-tween means and ends, is the argument based upon deterrence.The contention that the threat of execution will cause at leastsome killers to abandon their plans is intuitively appealing, andcontinues to draw wide support.95

the ultimate price, he purges his guilt and is reconciled to God. Apparently, theMormon doctrine of "blood atonement", which maintains that a murderer who hasspilled another's blood must not only die, but actually have his own blood spilled, inorder to expiate his guilt in the eyes of God was instrumental in leading the Utahlegislature to establish and maintain the firing squad as a mode of execution. SeeMartin R. Gardner, Illicit Legislative Motivation As a Sufficient Condition for Uncon-stitutionality Under the Establishment Clause-A Case for Consideration: The UtahFiring Squad, 1979 WASH. U. L.Q. 435, 440-49. This type of "rehabilitation," howev-er, is hardly what the term signifies in common usage; this type of "rehabilitation"actually seems more properly to be a form of retribution.

94. The public understandably is concerned with the possibility of a convictedmurderer being released and killing again, but the rate of recidivism among releasedmurderers is extremely low. See Hugo Adam Bedau, Recidivism, Parole and Deter-

rence, in THE DEATH PENALTY IN AMERICA 173, 175-80 (Hugo Adam Bedau ed., 3ded. 1982). This, of course, may be due to the fact that the least pernicious killersare the most likely to be paroled. Even this concern is mooted in large part by im-posing a sentence of life without parole.

95. The most prominent academic proponent of the existence of a deterrent effecthas been Isaac Ehrlich. See Isaac Ehrlich, The Deterrent Effect of Capital Punish-ment: A Question of Life and Death, AM. ECON. REv., June 1975, at 397; IsaacEhrlich, Deterrence: Evidence and Inference, 85 YALE L.J. 209 (1975). But seeZIMRING & HAWKINS, supra note 57, at 174-79 (attacking Ehrlich's views and meth-odology). The empirical evidence, however, probably is less important in the debatethan the strong intuitive feeling that somehow, capital punishment must deter. Er-

[Vol. 38:487502



The notion that deterrence stands as a sufficient justificationfor the death penalty is subject to two strong lines of attack.First, empirical support for the existence of a deterrent effect is,at best, sparse. No correlation exists between the presence orfrequency of the death penalty and the murder rate in variousstates." Within particular states, the murder rate does notseem to respond to changes in the imposition of death.97 Al-though the findings are not quite unanimous, 98 and cannot bedefinite because clinical precision cannot be obtained in the ab-sence of knowing the unknowable fact of how many murderswould have been committed at a particular time in a particularstate were the law different, the general consensus among socialscientists is that the deterrent effect of the death penalty isunproven.'

One response to this situation might be to argue that the bur-den is not on the defenders of a historically common punishmentto prove its efficacy, but rather upon its opponents to prove theopposite."® Because that probably cannot be done, given theinability to set up a valid controlled experiment, history andintuition should prevail. This proposition, however, does notanswer the second argument against the use of deterrence tojustify the death penalty. If deterrence alone could justify thedeath penalty, then there would be no reason to limit its use tocrimes involving the taking of life. Indeed, there would be no

nest van den Haag wrote: "Even though statistical demonstrations are not conclu-sive... I believe that capital punishment is likely to deter more than other pun-ishments because people fear death more than anything else.... Whatever peoplefear most is likely to deter most." Ernest van den Haag, The Death Penalty OnceMore, 18 U.C. DAVIS L. REV. 957, 965-66 (1985).

96. See Thorsten Sellin, Homicide in Retentionist and Abolitionist States, in CAPI-TAL PUNISHMENT 135 (Thorsten Sellin ed., 1967).

97. See SHIN, supra note 8, at 25-36.98. See Ehrlich's two articles, supra note 95.99. See ZIMEING & HAWKINS, supra note 57, at 167-86.

100. This appears to be the position taken by the Supreme Court:Considerations of federalism, as well as respect for the ability of a legis-lature to evaluate, in terms of its particular State, the moral consensusconcerning the death penalty and its social utility as a sanction, requireus to conclude, in the absence of more convincing evidence, that the in-fliction of death as a punishment for murder is not without justification.

Gregg v. Georgia, 428 U.S. 153, 186-87 (1976).

1997] 503



reason to limit its use to any category of the most seriouscrimes. Surely the risk of death might deter burglars, embez-zlers, those who operate illegal gambling operations, and a hostof others. Indeed, because many of these lesser crimes morecommonly are the products of calculated decisions to break thelaw than is homicide, which often is the product of an eruptionof passion, the use of the death penalty might be more of a de-terrent for these lesser crimes.'0 ' History validates this posi-tion; prior to the nineteenth century, executions for crimes thatneither took nor endangered life were common.'

Over the last two hundred years, governments clearly haverejected that conclusion, first, by legislatures drastically limitingthe list of capital crimes, and finally, by the Supreme Court rul-ing that the constitution demands some proportionality betweencrime and punishment and that therefore death may serve topunish only those who take, or at least threaten, life. 103 It isapparent, then, that deterrence alone, even if well established,cannot serve to fully justify the death penalty.

Perhaps because of this realization, but more likely because ofthe inability to demonstrate actual deterrent effects of capitalpunishment, proponents largely have shifted ground.", If notactually conceding deterrence's failure to justify capital punish-ment, academic proponents have come to rely more on retribu-tion to justify the death penalty.0 5 Commentators have at-tempted to demonstrate that retribution, properly understood, is

101. Thus, in Coker, the dissenters pointed out that the death penalty might welldeter rapists at least as effectively as it deters murderers. See Coker v. Georgia, 433U.S. 584, 617 (1977) (Burger, C.J., dissenting). The majority countered this view bystating that a penalty may not be disproportionate 'even though it may measurablyserve the legitimate ends of punishment." Id. at 593 n.4.102. The earliest Anglo-American abolitionist arguments were directed largely

against the broad range of capital offenses. For a list of dozens of capital crimesother than murder, see the excerpt from P. Colquhoun (1800), reprinted in 1MONTAGU, supra note 82, at 80-83.103. See supra notes 68-70 and accompanying text.104. See van den Haag, supra note 95, at 965.105. Thus, Ernest van den Haag, after pointing out that most abolitionists he has

discussed the question with would still abolish the death penalty even if its deter-rent effect could be proven, states: "[Deterrence] is not necessary for me either, sinceI would be for capital punishment on grounds of justice alone." Id. at 965.

504 [Vol. 38:487


1997] DEATH PENALTY AND MIMETIC VIOLENCE 505

a more sophisticated concept than mere revenge.' Unlike de-terrence, retribution depends heavily upon concepts of propor-tionality, and upon the idea that to call upon the criminal toexperience approximately the same fate as his victim is almostself-evidently just."'7 By committing the crime, the criminalhas, in effect, indicated that he believes that such an act is per-missible. When society inflicts the same fate upon the criminal,it then does so, in a sense, at the criminal's own invitation.'In a more metaphysical way, the retributive act restores somesort of balance that existed prior to the commission of thecrime.

0 9

Like deterrence, retribution cannot fully supply the underly-ing rationale for severe punishment. Most strikingly, the princi-ple that punishment literally should recapitulate the crime,which is at the heart of the retributivist case for capitalpunishment,"0 clearly would be rejected in a wide range ofother contexts. A person who is guilty of assault is not physicallyassaulted and a rapist is not raped. Little doubt exists that werea state to either assault the assaulter or rape the rapist, suchpunishment would be found unconstitutional."' The response

106. See, e.g., HENBERG, supra note 4, at 17-29.107. See id& at 59-76.108. This notion is associated most closely with Immanuel Kant.

The undeserved evil which any one commits on another, is to be regard-ed as perpetrated on himself. Hence it may be said: "If you slander an-other, you slander yourself; if you steal from another, you steal fromyourself; if you strike another, you strike yourself; if you kill another,you kill yourself."

ItMANUEL KANT, THE PHILOSOPHY OF LAW 196 (photo. reprint, Augustus M. KelleyPublishers 1974) (W. Hastie trans., 1887).109. Marvin Henberg has written about this "hydraulic" or "homeostatic" meta-

physical view, where God or nature demands "an equal and opposite reaction toeach crime or sin" in order to place the world back in balance. See HENBERG, supranote 4, at 81-83.110. See id. at 165.111. Even the dissenters in Furman could state: "Neither the Congress nor any

state legislature would today tolerate pillorying, branding, or cropping or nailing ofthe ears-punishments that were in existence during our colonial era. Should, how-ever, any such punishment be prescribed, the courts would certainly enjoin its execu-tion." Furman v. Georgia, 408 U.S. 238, 430 (1972) (Powell, J., dissenting). See alsoJackson v. Bishop, 404 F.2d 571 (8th Cir. 1968), where the court, in an opinionwritten by Judge, later Justice, Harry Blackmun, found Arkansas's use of the strapto discipjine prisoners to be unconstitutional. Id. at 579.


WILLIAM AND MARY LAW REVIEW [Vol. 38:487

to this argument might be that only in the case of homicide isliteral retribution necessary; in other cases, a rough equivalenceof pain can be struck through some combination of prison, mone-tary penalties, and shame to equal the harm caused."' Onlydeath, however, equals death.' The loss of all freedom, assets,and honor is inadequate.14

This view is problematic for a number of reasons. First, itproves too much. If homicide uniquely justifies capital punish-ment, then what of the long history .of use of the penalty forlesser crimes? Death was chosen as a punishment centuries be-fore it could be justified as a mirror image of the crime in-volved.'15 The justification of retribution seems to be a post hocrationalization, rather than the original foundation, of the deathpenalty. Second, along the same lines, if the punishment mustmirror the crime, should all willful homicides mandate the deathpenalty? Such a regime has been found unconstitutional,"6 andit almost certainly would meet with public disapproval. The exis-tence of the death penalty as a likely or certain punishment has,in the past, led juries to acquit even in the face of strong evi-dence of guilt."7 Popular opinion does not support the notion

112. See RICHARD POSNER, ECONOMIC ANALYSIS OF LAW 205-12 (2d ed. 1986) (stat-ing that the legal system generally acts in a way that indicates that the harm doneby crimes can be translated into a common currency, either of money itself or jailtime, sufficient to deter and to redress whatever harm was done).113.

[W]hoever has committed murder, must die. There is, in this case, nojuridicial substitute or surrogate, that can be given or taken for the sat-isfaction of Justice. There is no Likeness or proportion between Life,however painful, and Death; and therefore there is no Equality betweenthe crime of Murder and the retaliation of it but what is judicially ac-complished by the execution of the Criminal.

KANT, supra note 108, at 198.114. See id.115. See DAVID D. COOPER, THE LESSON OF THE SCAFFOLD (1974) (indicating that

the death penalty was widely used in the 1700s).116. See Woodson v. North Carolina, 428 U.S. 280 (1976).117. Judge Robert Sullivan, in examining the history of the trial of Lizzie Borden,

concluded that the fact that a conviction would have led to the death penalty was asignificant factor in leading to an acquittal despite copious evidence of guilt. SeeROBERT SULLIVAN, GOODBYE LIZZIE BORDEN 191-94 (1974). Sullivan went on to won-der "how many persons have escaped punishment for murder solely because thedeath penalty was the jury's only alternative to acquittal." Id. at 194.

506



that all murderers should die. Executing some murderers be-lieved to be particularly blameworthy should satisfy the retribu-tive urge.

Third, apart from the death penalty, retribution does not seemto be the primary goal of punishment, even if one accepts thenotion that the harm inflicted by the criminal can be translatedinto a roughly equivalent amount of loss of liberty. If retributionmotivated the punishment of all crimes, the public would notdemand increased prison terms for a wide range of crimes.When the public calls for more severe punishment for thieves orburglars or vandals, the nature of the crimes has not changed.What has changed is society's assessment of the efficacy of thepresent punishment as deterrence. A consistent retributive re-gime would serve to limit increases, as well as reductions, inpunishment.

Perhaps the practice rather than the theoretical underpin-nings of the death penalty present the most profound problemwith employing retribution as a justification. Retribution is, ofcourse, absolutely dependent upon the guilt of the criminal."'One powerful argument for retribution as a better theory thandeterrence is that deterrence might be achieved by the executionor other punishment of the innocent, as long as the general pop-ulation believed that they were guilty.' To the retributivist,however, inflicting a penalty, especially the death penalty, uponan innocent person would be inexcusable.

Among the general public, perhaps the most common citationin support of retribution is the Old Testament edict demanding

118. See BEAN, supra note 39, at 13 (discussing the essential connection betweenpunishment and guilt).119. Perhaps the easiest criticism of pure reliance on utilitarian theory to justify

punishment is that "it would be possible to manufacture evidence against an inno-cent man to set an example to others.... Alternatively. . . . if the advantage ofdeterrence could be achieved by seeming to punish the criminal it would then bepossible to pretend to punish him." Id. at 34. No one would defend intentional exe-cution of the innocent merely for deterrence, but some are untroubled by the execu-tion of the innocent by mistake, concluding that the social benefits of the death pen-alty outweigh "the loss of innocent lives through miscarriages." van den Haag, supranote 95, at 967.

19971 507



an eye for an eye. 2 ' This, like all retributionist theories, is adouble-edged command. It not only warns against inadequatepunishment, but also against excessive punishment. 2' Fur-ther, it warns even more strongly against punishment of theinnocent."= It is instructive to see how ancient Israel itself putthe demand into practice, especially with respect to the deathpenalty. The procedural demands necessary to sustain a capitalsentence were increased to a level that would put the WarrenCourt to shame." The Sanhedrin, the judicial body having fi-nal authority over death sentences, it is said, would be regardedas "bloody" or excessively prone to sustain death sentences ifexecutions exceeded one every seven years."

Observers have noted that this same phenomenon has becomemanifest in contemporary Israel. Not only is the death penaltyreserved for those implicated in genocide against the Jewishpeople," but the recent remarkable restraint of Israeli appel-late courts in freeing John Demjanjuk, accused of atrocities dur-

120.The first and, in most parts, oldest version of the lex talionis is Exod.21.22-25: "When men strive together, and hurt a woman with child, sothat there is a miscarriage, and yet no harm follows, the one who hurther shall be fined, according as the woman's husband shall lay upon him;and he shall pay as the judges determine. If any harm follows, then youshall give life for life, eye for eye, tooth for tooth, hand for hand, foot forfoot, burn for burn, wound for wound, stripe for stripe."

HENBERG, supra note 4, at 69 (quoting 21 Exodus 22:25).121.

Critics of retributive thought often argue that retribution is indistinguish-able from revenge, but it is exactly the search for limitation that sepa-rates the two. Retribution is a measured return of evil according to somenotion of what an agent (or group) is perceived to deserve. Revenge, onthe other hand, is an unmeasured return of evil that may or may notconnect to desert.

Id. at 18.122. See Irene Merker Rosenberg & Yale L. Rosenberg, Guilt: Henry Friendly Meets the

MaHaRal of Prague, 90 MICH. L. REv. 604, 618 (1991) (noting that Jewish law developed"rules ... so strict that they assureld] conviction of only the factually guilty").123. See id. at 616-25. The authors noted that, compared to Jewish law relating to

criminal procedure, "the restrictions embodied in the Bill of Rights [are] rather tep-id." Id. at 618.124. Or perhaps one every 70 years. See id. at 618 n.83.125. See TOM TEICHOLZ, THE TRIAL OF IVAN THE TERRIBLE: STATE OF ISRAEL V.

JOHN DEMJANJUK 96 (1990).

508



ing World War II, demonstrates the demands placed on one whowould seek retribution of the most drastic and final kind to takesimilarly extreme measures to assure that unjust executions donot occur."'

This same impulse can be seen in decades of American law.The unique nature of the death penalty has led the SupremeCourt to demand what some have called "super due process," aweb of procedural protections beyond those usually afforded tocriminal defendants.127 In recent years, however the notionthat capital cases require extraordinary procedures has lostmuch steam." Rather than seeing long delays and the infre-quent use of the death penalty as reaffirmation of the demandsof certainty associated with retributive theory, some see such de-lays as illegitimate.129

Perhaps most striking has been the Court's discussion of thequestion of actual innocence. To delay an execution because ofconstitutional violations collateral to guilt may well interferewith retribution; to delay an execution because of lingeringdoubt as to guilt is quite different. Yet the Court, although notwithout hesitation and dissent, has set formidable proceduralhurdles to those who claim new evidence of innocence.' TheCourt has refused to hold explicitly that the execution of a de-fendant in the face of significant evidence of innocence would beunconstitutional.'' If delay on collateral questions places retri-

126. See Iva L. Shafiroff, A Nazi Saved by Jews: Israeli Court Was Very Generousto Demjanjuk, LA DAILY J., Sept. 24, 1993, at 6. Demjanjuk's appeal was "one ofthe longest appeals in Israel's history." TEICHOLZ, supra note 125, at 302.127. See William S. Geimer, Death at Any Cost: A Critique of the Supreme Court's Re-

cent Retreat from Its Death Penalty Standards, 12 FLA. ST. U. L. REV. 737, 738 (1985).128. See id. passim; Yale L. Rosenberg, Kaddish for Federal Habeas Corpus, 59

GEO. WASH. L. REV. 362 (1991).129.

A pattern seems to be developing in capital cases of multiple review inwhich claims that could have been presented years ago are brought for-ward-often in a piecemeal fashion-only after the execution date is set orbecomes imminent. Federal courts should not continue to tolerate-even incapital cases-this type of abuse of the writ of habeas corpus.

Woodard v. Hutchins, 464 U.S. 377, 380 (1984) (Powell, J., concurring).130. See generally Joseph L. Hoffman & William J. Stuntz, Habeas After the Revolution,

1993 SUP. CT. REV. 65 (arguing that the Court in recent years has restricted the use ofhabeas because it does not think of habeas as part of the criminal justice system).131. In Herrera v. Collins, 560 U.S. 390 (1993), the Court rejected a second habeas

19971 509



bution beyond utilitarian concerns in a hierarchy of values, de-lay in the interest of resolving claims of innocence surely doesthe opposite."3 2

A strong adherent of deterrence theory would not find thistroubling; the existence of executions of the innocent is the pricepaid for the social benefit of deterrence.' Few deterrence pro-ponents, however, seem willing to go that far, at least explicit-ly."M When considering an individual case, surely deterrencealone may not justify the death penalty. Yet, when the analysisshifts to the system as a whole, deterrence is invoked moreprominently.'35 The effect of accepting deterrence as sole justi-

corpus petition by a death row inmate that was based on evidence of actual inno-cence, rather than on any alleged violation of procedural rights. See id. at 393, 418-19. Although three dissenters would hold that a defendant who could show "that heis probably innocent" is entitled to habeas corpus relief, id. at 442 (Blackmun, J.,with whom Stevens & Souter, JJ., join, dissenting), the majority disagreed. See id.at 402-03. Justices Scalia and Thomas simply would hold that there is no constitu-tional right to judicial consideration of newly discovered evidence. See id. at 427-29(Scalia, J., with whom Thomas, J., joins, concurring). The other Justices assumed,"for the sake of argument," that a "truly persuasive demonstration" of innocence,meeting an "extraordinarily high" standard of proof, might require federal interven-tion "if there were no state avenue open to process such a claim." Id. at 417. Be-cause the Court suggests that the existence of the governor's pardon power is anacceptable alternative "state avenue," see id. at 411-17, the suggestion seems to bethat the actual innocence argument is open only when a governor denies clemencyin the face of indisputable proof of innocence. In other words, it is unlikely ever tobe available in practice.132. "[B]ecause of the very disruptive effect that entertaining claims of actual inno-

cence would have on the need for finality in capital cases, and the enormous burdenthat having to retry cases based on often stale evidence would place on the States,"the Court has rejected the notion that actual innocence claims should be routinelyheard. Id. at 417.133. See, e.g., van den Haag, supra note 95, at 967 (rejecting the notion that mis-

carriages of justice support abolishing the death penalty). In contrast, Margaret JaneRadin has written:

A pure utilitarian would argue that we should execute whomever, andhowever many, we need to in order to deter the "right" amount ....Whether someone is guilty of a crime or deserves to die for it is not ofconcern to the pure utilitarian. But no one in her right mind is a pureutilitarian.

Margaret Jane Radin, Proportionality, Subjectivity and Tragedy, 18 U.C. DAVIS L.REV. 1165, 1170 (1985).134. See, e.g., Radin, supra note 133, at 1170.135. See Mary Ellen Gale, Retribution, Punishment, and Death, 18 U.C. DAVIS L.

REV. 973, 1019 (1985).

510



fication for systemic imperfections is, of course, to accept, per-haps unconsciously, that guilt is not essential.'36 Such a con-clusion undermines any sophisticated retributive justification forpunishment.

Thus, neither of the commonly invoked rationales succeeds inexplaining the acceptability of the current system of capital pun-ishment. Deterrence clearly fails, whether examined with an eyetoward the empirical question of whether it is effective, or thetheoretical question of whether, if effective, it alone can justifycapital punishment. Retribution can serve as a rationale for cap-ital punishment, but it cannot explain the willingness of a sys-tem of capital punishment to accept any noticeable likelihood ofthe execution of the innocent. Likewise, it is hard to see how re-tributive theory supports a system in which only a tiny subset ofkillers is put to death, and those who are, are not necessarilythe most culpable." 7

One is left, then, with two alternatives. The first alternative isthat the death penalty successfully is defended by the deft use ofboth deterrence and retribution in tandem. Thus, retributionwould justify most executions and those that cannot be so justi-fied would instead be justified by deterrence. One or the othertheory would succeed in explaining each aspect of contemporaryAmerican capital punishment. Indeed, this often seems to be theway in which the death penalty actually is defended, whetherconsciously or not. 3 This explanation is unsatisfying. It seems

136. Cf van den Haag, supra note 95, at 967 (explaining that imperfections areinherent in any government-operated system and that innocents are killed accidental-ly by routine government functions).137. A sophisticated retributive theory does not focus merely on the act; it also

focuses on "the subjective culpability of the offender committing it." Gale, supra note135, at 1012. Clearly, the Supreme Court's post-Gregg insistence on some consider-ation of the particular circumstances of the murder recognizes this principle. SeeGregg v. Georgia, 428 U.S. 153, 183-87 (1976) (requiring that the blameworthiness ofthe defendant be considered). Reaction to the imperfection of "proportionality review"procedures varies. Compare F. Patrick Hubbard, 'Reasonable Levels of Arbitrariness"in Death Sentencing Patterns: A Tragic Perspective on Capital Punishment, 18 U.C.DAVIS L. REV. 1113 (1985) (arguing that no system can be perfect and that we mustaccept a level of "reasonable arbitrariness"); with Radin, supra note 133, at 1172(concluding that "no one can deserve to die when the issue of whether anyone can(ideally) deserve to die is morally uncertain and when in any event it is certain thatwe cannot select out all and only those who deserve to die").138. Cf Hardy Jones & Nelson Potter, Deterrence, Retribution, Denunciation and

1997] 511



too cobbled together, too obviously an after-the-fact attempt todefend an a priori position.

The second alternative is that neither of the rationalist expla-nations is really at the core of society's acceptance, perhapsneed,'39 for capital punishment. At least occasionally, observersnote the symbolic uses of capital punishment. " ° Often, this isdone in a way that emphasizes the entertainment value of anexecution; however, this may also be subsumed in deterrencetheory, because the symbolism of the execution is seen as awarning to others, or in retributive theory, because that symbol-ism is meant to express the community's outrage.'4 Inevitably,then, discussions of symbolism collapse back into discussions ofone of the major rationalist justifications, or they simply assumethat the symbolism of the act, once removed from rational expla-nations, self-evidently cannot support the death penalty.

Perhaps, however, this argument dismisses too easily the non-rational, not merely as an explanation of what goes on, but moreimportantly, as something that has such strength that it mustbe dealt with in a more respectful way rather than simply beingdismissed as unworthy. A possible place to start such an effortto understand and to react to the power of the death penalty canbe found in the work of the anthropologist and social and liter-ary critic Ren6 Girard.

Death Penalty, 49 UMKC L. REV. 158, 169 (1981) (arguing that even if capital punishmentcould be based on a deterrence theory, the justification would have to be supported byretributive consideration).139. Cf HENBERG, supra note 4, at 32-42 (theorizing that humans have biologically

and culturally evolved to desire retribution for past wrongs).140. See, e.g., Gale, supra note 135, at 996-99; Jones & Potter, supra note 138, at

167-69 (arguing that the death penalty is expressive, but in the opposite way thanits defenders believe); see also, e.g., JOEL FEINBERG, DOING AND DESERVING 96-105(1970) (discussing the expressive function of punishment generally); Elizabeth D.Purdum & J. Anthony Paredes, Rituals of Death: Capital Punishment and HumanSacrifice, in FACING THE DEATH PENALTY: ESSAYS ON A CRUEL AND UNUSUAL PUN-ISHMENT 139-55 (Michael L. Radelet ed., 1989) (exploring the "socio-religious" aspectsof executions by comparing modem executions in Florida with human sacrifices ofthe Aztecs).141. Thus, Professor Gale has classified denunciation as one of the utilitarian as-

pects of capital punishment. Gale, supra note 135, at 996-99. Purdum and Paredesview some explanations of the capital punishment ritual as being "unabashedly reli-gious." Purdum & Parades, supra note 140, at 151.

[Vol. 38:487512



III. MIMESIS AND THE SCAPEGOAT:THE THEORIES OF RENl GIRARD

Ren6 Girard, Professor of French Literature, Language, andCivilization at Stanford University, has developed a sweepingand controversial theory of culture."' Beginning with insightsdrawn from his original academic base in literary criticism, histheory has grown to include work in anthropology, religion, psy-chology, and other social sciences. As might be expected, thewide scope of his work invites criticism from specialists; the at-tempt to forge a simple unifying theory to explain human cul-ture will almost invariably claim too much. Yet one need notaccept the full measure of his thought to be impressed by thepower of his insights and their ability to illuminate at least sig-nificant parts of culture.

Beginning with his work as a literary critic, Girard exploredclassic portrayals of human action and, more significantly, themotivations and decisions behind them. Repeatedly, hefound that great works of literature demonstrate that the sourceof human desire is neither some characteristic of the person whodesires nor the object that is desired.'" Rather, desire is rootedin mimesis, the need to imitate others.'45 People want an ob-ject, a way of life, or the love of another person primarily be-cause someone else does as well. 45 This desire is the source ofenvy and recurring conflict. On occasion, as in Cervantes's DonQuixote, this compulsion to imitate can be comic. 47 Perhapsmore often, however, it leads to rivalry and possible tragedy.148

142. The most thorough presentations of Girard's ideas can be found in GIRARD, su-pra note 14, and REN9 GIRARD, THINGS HIDDEN SINCE THE FOUNDATION OF THEWORLD (Stephen Bann & Michael Metteer trans., 1987) [hereinafter THINGS HIDDEN].Probably the best single volume explaining and developing Girard's thought, particu-larly its implications for contemporary religion, is GIL BAILIE, VIOLENCE UNVEILED:HUMAN=TY AT THE CROSSROADS (1995).143. See Paul Dumouchel, Introduction, in VIOLENCE AND TRUTH: ON THE WORK OF

REN9 GIRARD 1-21 (Paul Dumouchel ed., 1988).144. See GIRARD, supra note 14, at 145.145. See iad at 143-68; THINGS HIDDEN, supra note 142, at 283-98; see also BAILIE,

supra note 142, at 111-20.146. See GIRARD, supra note 14, at 145.147. See THINGS HIDDEN, supra note 142, at 15-16; Dumouchel, supra note 144, at 3-6.148. This pattern of mimesis leading to rivalry and tragedy is a recurring literary

1997] 513



Girard did not stop at the insight that mimesis is often at thecenter of fiction, however. Examining material from the socialsciences, particularly anthropology, he concluded that humanshave a universal tendency toward mimesis. 49 At some level,this seems empirically irrefutable; the socialization of the child,of course depends on it. 5' Girard maintains that the imitativeimpulse continues throughout life, and unlike the situation in-volving the small child, often is hidden and unrecognized. 5' Toa certain extent, imitation is a positive occurrence, as one seeksto model oneself after others in ways that may be benign andeven beneficial. At some point, however, imitation becomes rival-ry and rivalry becomes hostile.'52 The desire to be like anotherperson, and to desire what the other person desires, leads toconflict, which itself leads to a redoubling of the effort to imi-tate, supplant, and become the rival.' Inevitably, this cyclemust lead to violence; the goal of cultural institutions is to inter-vene to stop it. 54

theme beginning with Greek tragedy, see GIRARD, supra note 14, at 46-48, and theearliest Bible stories, see BAILIE, supra note 142, at 137-40, and continuing throughthe works of Shakespeare, see THINGS HIDDEN, supra note 142, at 36, Dostoevsky,see id. at 338-47, and George Eliot, see Dumouchel, supra note 143, at 6-8.149. See THINGS HIDDEN, supra note 142, at 3-138.150.

[T]here is nothing, or next to nothing, in human behavior that is notlearned, and all learning is based on imitation. If human beings suddenlyceased imitating, all forms of culture would vanish. Neurologists remindus frequently that the human brain is an enormous imitating ma-chine....

... [I]n children-as in animals-the existence of acquisitive imitationhas been recognized by researchers....

An equivalent situation rarely occurs among adults. That does notmean that mimetic rivalry no longer exists among them ... butadults . . . have learned to fear and repress rivalry.

Id. at 7, 9.151. See id. at 7-19.152. See id.153. Thus, the function of the law, and other cultural institutions, is to prohibit

imitation, at least that sort of imitation that spirals out of control and becomes de-structive. See id. at 10-19. This function is inherently very tricky. Much imitationgives rise to cohesion, and, of course, to cultural transmission. See id. To this extent,mimesis is necessary. At some point, however, it becomes destructive. See i& Lawand other cultural institutions attempt to stop it at or prior to that point. See id.154. See id.

514


19971 DEATH PENALTY AND MIMETIC VIOLENCE 515

Normal cultural institutions, however, may inadequately re-strain this escalating violence because mimesis ensures thateach violent act attracts its own imitators, and each mimeticdesire converges so that more and more individuals see them-selves as rivals for the same objects."s At some point, the needto stop the cycle of rivalry requires that a dramatic step be tak-en to reunite the community. Because violence must return vio-lence, this step itself must be violent.'56 A common enemymust be identified, and the community must come to believethat this enemy is the source of all discord in the communi-ty.'57 The enemy then can be dispatched and this act of vio-lence will not lead to a response, because the victim will have nosupporters in the community. At least until the cycle repeats it-self, society averts the crisis caused by mimetic violence.'

A huge paradox then arises. The victim who was targeted asthe source of all evil now has become the source of thecommunity's peace and order."9 The victim in some way nowbecomes sacred, the bestower of good and evil. 60 Perhaps the

155. See GTRARD, supra note 14, at 18-27.156. "Only violence can put an end to violence, and that is why violence is self-

propagating. Everyone wants to strike the last blow, and reprisal can thus followreprisal without any true conclusion ever being reached." Id. at 26.157.

[A]ny community that has fallen prey to violence or has been stricken bysome overwhelming" catastrophe hurls itself blindly into the search for ascapegoat. Its members instinctively seek an immediate and violent curefor the onslaught of unbearable violence and strive desperately to con-vince themselves that all their ills are the fault of a lone individual whocan be easily disposed of.

Id. at 79-80.158.

Where only shortly before a thousand individual conflicts had raged un-checked between a thousand enemy brothers, there now reappears a truecommunity, united in its hatred for one alone of its number. All therancors scattered at random among the divergent individuals, all the dif-fering antagonisms, now converge on an isolated and unique figure, thesurrogate victim.

Id. at 79.159. "[Tlhe surrogate victim-or, more simply, the final victim-inevitably appears

as a being who submits to violence without provoking a reprisal; a supernatural be-ing who sows violence to reap peace; a mysterious savior who visits affliction onmankind in order subsequently to restore it to good health." Id. at 86.160. See id.



community now will see the victim as having controlled theentire process, rather than having been its passive object.16' Inthis process, Girard saw the origin of religion.

According to Girard, religion attempts to recreate and main-tain the sense of the community's unity through controlled reen-actment of the murder of the victim.'62 Perhaps through ahighly ritualized actual killing, perhaps through some symbol-ism or reenactment of gestures short of the killing itself, com-munity leaders will repeat this founding murder." At thesame time, however, the reenactment will be strictly limited. Itwill be performed only when, by whom, and in the way so pre-scribed by the community. To engage in the same act outside ofthe sanctioned parameters will be prohibited, will be, in effect,sacrilege.'6

The victim, of" course, need not actually be guilty of anythingin order for the process to be effective. In fact, the victim mostassuredly is not guilty of all that the community believes him tobe guilty of-that is, instigating all of the community's discord.To bolster his argument, Girard relied on the practice of primi-tive societies lacking an accepted and powerful central governingsystem.'65 When a murder threatens the stability of the com-munity, the community's leaders will find a victim to sacrifice,but quite consciously will avoid selecting the person who actual-ly is guilty of the act that threatens the community.'66 Al-though this approach appears to be bizarre, it is quite logicalwhen one recalls that the fundamental purpose of the sacrifice isto bring peace to the community.6 7

The true killer likely is a full member of the community, onewith relatives and friends. If he is killed, it will set off anotherround of retaliation from those who will take his side. Instead,

161. See id. at 94-99.162. See id. at 99.163. Thus, the function of ritual is to require imitation of unifying violence, but

under strict control. See THINGS HIDDEN, supra note 142, at 19-23.164. In addition to the required reenactments of religious ritual, the primary func-

tion of law and other cultural institutions is to restrain mimesis. See id. at 10-19.165. See GIRARD, supra note 14, at 22.166. See id. at 21-22.167. See id.

[Vol. 38:487516



the community leaders will select one who can deflect the cycleof violence by assuming the role of the enemy of the entire com-munity, including the guilty party's clan.1" Although the vic-tim has been chosen arbitrarily, the community will actually be-lieve that he is responsible for the community's discord.'69 Todestroy this belief exposes the killing of the victim as merelyanother step in the cycle of violence that would itself requirecounterviolence.

Girard has noted that the sacrificial victim is seen as both apart of, and outside, the community. 7 He is part of the com-munity in that his death will bring peace, and he will, therefore,be seen as intending benefit to the community.' 7' The sacrificecomes about by directing the entire community against the vic-tim; labeling him as distinctly "sacrificable."172 Examples ofthis behavior in primitive tribes are not hard to find. Often thesacrificial victim is a stranger or outsider, but not until thecommunity somehow virtually enfolds him.73 When animalsacrifice replaces human sacrifice, it is likely to be the sacrificeof a domestic animal, that is, one who lives within the humancommunity.

174

Although the victim may have been chosen arbitrarily, the

168. In a modem culture, the way to gain unanimity is for the system to "reorga-nize itself around the accused and the concept of guilt.... Instead of following theexample of religion and attempting to forestall acts of revenge ... or to redirectthem to secondary objects, our judicial system rationalizes revenge . . . ." Id. at 21-22. The purpose, however, remains the same, to unite all in the community and toend the cycle of violence. See id.169. See id.170. '[AilI victims, even the animal ones, bear a certain resemblance to the object

they replace; otherwise the violent impulse would remain unsatisfied. But this re-semblance must not be carried to the extreme of complete assimilation, or it wouldlead to disastrous confusion." Id. at 11.171. See id. at 270-71.172. Id. at 270-72 (The victim becomes a "monster.. . no longer regarded in the

same way as the other members of the community").173. Thus, prisoners of war and slaves commonly were chosen as sacrificial victims,

as were children. See id. at 271. In all those cases, the victim is not fully part ofthe community, but is nevertheless within it. See id. In a more advanced society,judicial condemnation will serve to brand one as being beyond the moral community,while still living within it.174. See id. at 272-73.

1997] 517



community still must believe in his guilt.'75 Primitive societieshave developed rituals in which the sacrificial victim is encour-aged, or even required, to violate social taboos, and to claimgreat privileges prior to the time of the sacrifice.76 Thus, thecommunity sees the sacrifice as avenging these acts, and reaf-firms its adherence to the taboos that serve to restrain the ac-quisitive rivalry that exists in community life.'77

The need to re-create, at least symbolically, the "foundingmurder" may be seen in a number of institutions. Girard sawthis as the basis of all religions.' The pattern may be seenelsewhere, however, especially in a time in which traditionalreligion receives less community respect. The pattern by which,for example, entertainers or politicians are singled out as spe-cial, showered with privileges and praise, and then attacked bythe same people who lionized them, symbolically, if not literally,"killing" them by destroying their career or reputation, seemsconsistent with the victimage process. The contemporary appli-cation of the death penalty is also consistent with this process.

To hypothesize that the primary reason behind the death pen-alty is its function as a symbolic sacrifice, meant to bring aboutpeace by uniting the community, permits one to explain a num-ber of features of the contemporary death penalty that simplyare not consistent with the rationalist theories that either deter-rence or retribution serve as its basis. Some of these featuresare peripheral, for example, the tradition of the last meal. Thistradition is akin to the traditions in primitive cultures, wherebythe sacrificial victim is permitted, if not required, to exerciseprivileged behavior prior to the sacrifice.'79 The ritualized priv-ilege gives the community a reason to-kill because the victimhas claimed a privilege that he does not deserve, but also marks

175. See id. at 271-72.176. See id. at 104-07.177. In some tribes a prisoner chosen to be killed would first be "encouraged to

violate the laws," permitted to steal, or subjected to a ritualistic attempt to escape.See id. at 274-75. Societies might choose to execute their kings, and require that theking ritually engage in forbidden acts, including acts of violence or incest, eitheractually or symbolically. See id at 104-10.178. See THINGS HIDDEN, supra note 142, at 30-47.179. See GMARD, supra note 14, at 275.

[Vol. 38:487518



the victim as, paradoxically, someone special and worthy ofspecial treatment.'80 Thus, the condemned prisoner may haveanything he wants as a last meal; once again privilege precedesexecution.

More importantly, Girard's theories explain more salient fea-tures of the death penalty. For example, if retribution or deter-rence were truly the foundation of the contemporary death pen-alty, one would expect that it would be imposed on the most cal-culating, evil killers. Yet one sees that, instead, it is often im-posed on those who, like the sacrificial victim, are like typicalmembers of the community, but at the same time are significant-ly different. Executing the young,'' the mentally retarded,8 'or members of a minority group'" will not serve deterrence andin many cases will not serve retribution, as well as would execut-ing those who more closely resemble the majority of the commu-nity. As we have seen, the choice of a victim who is in some sig-nificant way not like everyone else actually furthers the efficacyof the sacrificial victim by making it easier for most, or all, of thecommunity to avoid empathy or identification with him.'"

The community's fascination with at least some of those exe-cuted individuals also comports with Girard's theories. Why doesa Gary Gilmore or a Ted Bundy continue to be the subject ofbiography and even fiction years after he is dead? One might saythat it is a function of society's "fascination with evil," but thatphrase merely restates, rather than explains, the phenomenon.Could this be a reflection of the phenomenon of elevating thesacrificial victim to the status of a god?'"

If the function of the sacrifice is to unify a fragmented com-munity, and if ritual killing rather than actual killing can per-form that function, one would expect actual killing to be in de-mand most where ritual is least effective. Thus, where religion

180. See id.181. See supra note 75 and accompanying text.182. See supra note 76 and accompanying text.183. See supra note 79 and accompanying text. But see infra text accompanying

notes 206-08 (discussing the impact of the belief that the death penalty isdisproportionately applied to minority groups).184. See supra notes 169-73 and accompanying text.185. See GIRARD, supra note 14, at 275.

1997] 519



is more privatized, and where religious and other social diversityexists, the legal system and its rituals may need to assume thisreligious function. One would predict, then, that the death pen-alty most likely would be used in a more culturally diverse soci-ety. This is consistent with the often-noted fact that the cultur-ally and religiously diverse United States is unique among west-ern democracies in its use of the death penalty." Nations thathave renounced the death penalty generally are more homoge-nous; such homogeneity provides alternative channels for ex-pressing community unity.'87

Most significantly, the theory accounts for the recent tendencyof courts to place less emphasis on complete certainty of guiltand more emphasis on expediting the execution process." Asthis Article has shown, the efficacy of the sacrifice depends notupon the actual guilt of the sacrificial victim,'89 but upon thecommunity's willingness to believe in that guilt. At some point,however, too cavalier an attitude toward guilt or innocencewould destroy the execution's unifying power; the executionwould merely become another act of unjust violence, calling forits own retribution, or as Girard would see it, imitation."9

Girard saw an evolution in the ways that society "avoid[s] be-ing caught up in an interminable round of revenge."' Thehighest level of human development is the creation of an effec-tive modern judicial system.92 Such a system constrains socialviolence by appealing to principles of justice, and identifying thecommunity's enemies by a convincing demonstration of guilt. 93

This emphasis on a reliable finding of guilt will, to many, sharp-ly distinguish modern legal systems from primitive societies that

186. See ZRMRING & HAWKINS, supra note 57, at 3-25.187. Of course, not all ways of celebrating cultural unity and homogeneity are posi-tive. Solidarity within the group can be the cause, or the effect, of xenophobia orwar. See BAILIE, supra note 142, at 60-64, 157-63.188. See supra notes 128-33 and accompanying text.189. See supra notes 166-67 and accompanying text.190. The judicial system can use its own violence to cure the violence of the crimi-

nal "without fear of contagion" only to the extent that it can rationalize revenge asbeing directed only at the guilty. GIRARD, supra note 14, at 21-23.191. Id. at 20.192. See id. at 20-21.193. See id. at 22.

520 [Vol. 38:487



relied on ritual sacrifice or such methods as trial-by-combat. ToGirard, these differences only mask the extent to which eachsystem serves the same goal.1" Like more primitive systems,modern legal systems seek to break the cycle of imitative vio-lence by directing the punitive urge of all members of societytoward a common enemy.

If the primary function of the death penalty is to provide aritual by which to unite the community in common opposition toa particular individual, allowing it to break the escalating cycleof intracommunity violence, what does that mean for the debateover the legitimacy of the death penalty? Are abolitionists ask-ing the right questions? Is effective argument even possible un-der these circumstances? How does a legal system based uponrational argument deal with a matter so weighted down withhidden, seemingly nonrational considerations?

IV. THE IMPLICATIONS OF GIRARD'S THEORIES TO THE DEATHPENALTY DEBATE

As an overarching explanation of all cultural institutions,Girard's theories of mimesis and sacred violence surely may besubject to criticism. Can all desire essentially be mimetic? Atsome basic level, the desirability of an object such as food is gen-erated by the self (without it, one dies) and by the nature of theobject (when one is hungry, the relative value of bread and card-board is surely not a function of the fact that someone elsechooses bread). Further, all modeling does not lead to escalatingrivalry. Some degree of imitation is absolutely essential to pre-serve the species; the young must model themselves in waysthat allow them to grow and thrive.

It is just as difficult, however, to deny that Girard has identi-fied phenomena that are of great significance and that explainmuch of human violence and the attempts, themselves violent,made to contain it. Specifically, Girard's theories have remark-able explanatory power when applied to the imposition of thedeath penalty. Much of what surrounds the death penalty andseems puzzling in light of rational analysis is utterly logical

194. See id. at 23.

1997] 521



when executions are seen as sacred mimetic violence.

A. How Well Does the Theory Explain the Use of CapitalPunishment in Contemporary America?

First, this Article addresses some peripheral matters. The tra-dition of the sumptuous last meal offered to the condemned pris-oner remains a subject of peculiar fascination for the media andthe public.'95 The ritualistic sacrifice performed by a number ofsocieties required that they grant some privileged status to thevictim prior to his execution.9 ' Thus, the victim might be in-vested with the trappings of royalty ceremoniously, and actuallyallowed to enjoy unusual benefits prior to the sacrifice. The "lastmeal" would seem to be a vestige of this type of ritualistic grantof privilege to one who will then be executed for the effrontery ofenjoying unusual privilege.

A truly bizarre, if uncommon, twist on capital punishment isprovided by instances in which a death row prisoner, perhaps onthe very brink of execution, falls ill or attempts suicide.'97 Atthis point, the state frantically works to restore the prisoner'shealth, so that he might then be executed.9 8 If the purpose of

195. Media descriptions of last meals are legion. See, e.g., Rhonda Cook, ExecutionIs Still on Schedule for Killer of Students in 1981, ATLANTA J. & CONST., May 2,1996, at C3, available in 1996 WL 8206111 ("fried bass, cole slaw, hushpuppies,french fries with ketchup, tartar sauce and hot sauce [and] either sweetened iced teaor a Coke"); Richard C. Paddock, Triple Killer Prepares to Die As Appeal Rejected,L.A. TIMES, May 3, 1996, at Al ("fried pork chops, baked potato, asparagus, saladwith bleu cheese dressing, French bread with butter, apple pie with ice cream and aglass of whole milk"); James Varney, Time Runs Out for Killer; Antonio James IsPut to Death, TIMES-PICAYUNE (New Orleans), Mar. 1, 1996, at Al, available inLEXIS, News Library, Notpic File ("fried oysters, fried shrimp, fried catfish, crawfishetouffee, seafood gumbo, hush puppies, French fries, pecan pie and Coca-Cola"). Infact, this phenomenon is not limited to the United States. See, e.g., Tan Ooi Boon,Martin Hanged, Leaving Behind Mystery Over Another "Victim", STRAITS TIMES (Sin-gapore), Apr. 20, 1996, at Home 25, available in LEXIS, News Library, Strait File("pizza and a cup of hot chocolate").196. See supra note 179 and accompanying text.197. Perhaps the most famous instance of this phenomenon involved Nazi leader

Hermann Goering. See JOSEPH E. PERSICO, NUREMBERG: INFAMY ON TRIAL 421-25,445-47 (1994). In some jurisdictions, more death row inmates die of suicide than areexecuted. See Michelle Locke, Suicide, Natural Causes: How Most Inmates Die onDeath Rows, CI. TRiM., Mar. 6, 1996, at 2, available in 1996 WL 2650108.198. Indeed, failure to do so might well violate the prisoner's right to medical

522 [Vol. 38:487



capital punishment is simply to achieve the rational benefitspresumed to flow from the prisoner's death, this sequence ofevents is absurd. If one examines ritual sacrifice, one sees thatsocieties place overwhelming importance on precise adherence tothe ritual; if each step is not followed exactly, it will not be seenas effective in re-creating the unifying primal sacrifice. 9 ' Theprisoner who commits suicide spoils the ritual. If the executionis seen as symbolic, it becomes clear why merely seeing to itthat the condemned prisoner is dead will not do.

More important is the explanatory power of Girard's theorieswith respect to more fundamental matters. If, on the one hand,the purpose of the death penalty is deterrence or retribution, onewould expect it to be seen as clearly inappropriate when some-one whom the community could not see as being fully volitionalor fully like others were involved. If, on the other hand, the pur-pose of the death penalty is to unite the community in a ritualexpression of violence against someone seen as an outsider, thena fully volitional requirement is inapposite; it actually furthersthe goal of community catharsis to execute someone who is no-ticeably unlike everyone else. The latter conclusion seems closerto describing the contemporary American death penalty. Youthand mental retardation, except when they become extreme, donot bar execution."'0 The media dwells consistently on the bru-tal, "animal-like" behavior of the condemned which led to hissentence.2"' Most commentators agree that Susan Smith wasspared the death penalty by the successful efforts of her lawyersto "humanize" her, by showing the jury that she was, ultimately,a person not that different from themselves, rather than someMedea-like, °20 alien, creature.0 '

treatment. See Estelle v. Gamble, 429 U.S. 97 (1976). In general, prison authoritieshave a duty to avoid "deliberate indifference" to prison suicides. See Tittle v. Jeffer-son County Comm'n, 10 F.3d 1535, 1539 (11th Cir. 1994).199. For a comparison of ritual in Aztec human sacrifice and the contemporary

American death penalty, see Purdum & Paredes, supra note 140, at 140-53.200. See supra notes 75-77 and accompanying text.201. See, e.g., Joseph J. Gilbert, Strange Justice in Maryland, WASH. POST, Oct. 2,1996, at A17 (noting, in an opinion piece, the brutal nature of a crime); WilliamMiller, Pirtle's Execution Upheld, Court Affirms Death Penalty for Brutal SpokaneMurders, SPOKESIAN REV. (Spokane, Wash.), Oct. 13, 1995, at B1, available in 1995WL 8790446.202. See generally, EURIPIDES, MEDEA, in 1 PLAYS OF EURIPIDES (Edward P.

1997] 523



The theory would seem to explain much of the media coverageof executions and those executed. Organized demonstrations out-side prison walls to celebrate the execution, and the media's cov-erage of these events, can be seen as attempts to formally pro-claim the community's newfound unity in its hatred of the crimi-nal.2" The continued fascination with the criminal, whetherGary Gilmore, John Wayne Gacy, or Ted Bundy, betrays some-thing more; the community makes the condemned man a celebri-ty, someone special, someone who has played an active role inuniting the community, albeit against himself.

Perhaps most significantly, Girard's theories help to explainwhy most of the capital punishment debate has been so fruitless.An argument based upon deterrence is beside the point if whatsupports the impulse behind executions is far removed from therational calculus of cost and benefit that the deterrence debateassumes as its foundation. An argument based upon retributionand desert likewise will miss the central point of the ritual thatsurrounds the death penalty. Additionally, the theory can ex-plain why the reinstatement of the death penalty has not endeddemands for its expansion and for other increases in the level ofpunishment of criminals.

Girard has stressed that the efficacy of the ritual killing of thecommon enemy is only effective if it actually does lead to sub-stantial unanimity among the rest of the community. °5 If asubstantial part of the community sees a killing as being unjust,it becomes just one more step in the mimetic escalation of vio-lence, one that itself calls for a violent response, which will in

Coleridge trans., London, G. Bell 1891) (involving a Greek enchantress who killedher children).203. See Naomi Chase, Acknowledging the Susan Smith in Us All, PLAIN DEALER

(Cleveland), Aug. 8, 1995, at 5E, available in 1996 WL 7124234; Mike Doming, Su-san Smith's Jurors Felt Her Pain: Quiet End Sought to Public Ordeal, CmI. TaM.,July 30, 1995, at 3.204. In fact, for many years, executions were public events, permitting the entire com-

munity to participate, at least vicariously. Eventually, however, public hangings lostsupport because crowds became more rowdy and officials no longer believed that thespectacle was having its desired deterrent effect. For a thorough recounting of the de-cline and eventual abolition of public executions see generally COOPER, supra note 115.205. "The true 'scapegoats' are those whom men have never recognized as such, in

whose guilt they have an unshaken belief." THINGS HIDDEN, supra note 142, at 46-47.

524 [Vol. 38:487



turn trigger its own countermeasures. To a certain extent, theUnited States already may have reached that state of affairs.Although polls show that large majorities favor the death penal-ty,..6 these majorities do not reach the point of consensus. Per-haps most disturbingly, to the extent that the death penalty isseen as disproportionately applied to members of identifiableminority groups,' °7 the tendency of the penalty to divide, rath-er than unite, the community becomes more acute.

If the ritual does not work, that is, does not unite the commu-nity, the reaction will not be to abandon it, but to insist on itsrepetition." 8 The attempts to do it just right, and therefore ef-fectively, will become more frantic. Here one reaches a criticalpoint in the analysis. If Girard's theories do, in fact, explain theessential nature and function of the death penalty, how shouldthe abolition debate be framed? Can the abolitionist argumentsucceed without removing something critical to social stability?

B. How Does the Theory Alter the Death Penalty Debate?

As has been previously discussed the death penalty debatehas taken shape around the central notions of deterrence andretribution.0 9 Although some commentators have noted the rit-ual aspects of capital punishment, these observations are usual-ly somewhat peripheral to the analysis.21 If Girard's theoriesare valid, the traditional framework for debating the death pen-alty must be altered. If the central function of the death penaltyis neither to deter potential bad actors by raising the cost of fu-ture killings, nor to punish on the basis of desert, but to rituallyunite the community through an act of common violence againsta designated adversary, the usual arguments for and againstcapital punishment may lose much of their relevance.

Girard himself will be of little help here. He does not use his

206. See TUSHNET, supra note 6, at 119-20.207. See supra notes 78-79 and accompanying text.208. See supra note 170 and accompanying text; see also Laurence A. Grayer, Com-

ment, A Paradox: Death Penalty Flourishes in U.S. While Declining Worldwide, 23DENY. J. INTL L. & POL', 555 (1995) (noting that only in the United States is theuse of the death penalty increasing).209. See supra Part II.210. See supra note 140.

19971 525



theories to make specific recommendations for social change.Indeed, he has cautioned against the use of his ideas as weaponsto critique specific social practices. 1' Instead, he has advocateda thorough societal rejection of mimetic violence, which he seesas the inevitable result of the growth of the indictment of vio-lence contained in the Christian gospels.212 Unlike Girard, le-gal commentators are required to deal with particular questionsof social policy, and are constrained from basing argumentsentirely upon religious faith.21 Still, if the death penalty is es-sentially an exercise in ritualized sacrifice meant to unify thecommunity, a number of specific consequences do seem to follow.Abolitionists will likely find Girard's theories more attractivethan retentionists, but even retentionists may find that they candraw support from the new analytical approach.

1. Implications for Retentionists

If Girard's theories are valid, the -symbolic function of thedeath penalty is not merely an interesting side issue, but centralto its existence. Girard maintains that the ritualized communityviolence employed by society is not merely a luxury, but servesan important purpose.214 It creates a mechanism throughwhich the escalating urge to imitate violence can be broken.1 5

If that mechanism is taken away, what will ensue? Girardsounds a warning: Effective rituals of violence, he says, appearto be essential to social cohesion. In their absence, imitativeviolence escalates unchecked.1 6

For retentionists, this could lead to a new line of argument.The argument would resemble classical deterrence arguments in

211. "What is interesting in our work here is not the possibility of making impres-sionistic applications of the theory in order to denounce any aspect of society weplease." THINGS HIDDEN, supra note 142, at 34.212. See id. at 180-223.213. For a discussion of the extent to which public policy arguments may accept-

ably rest on religious grounds in a liberal democracy, see KENT GREENWALT, RELI-GIOUS CONVICTIONS AND POLITICAL CHOICE (1988).214. See GIRARD, supra note 14, at 17 (stating that sacrifice by society is "an in-

strument of prevention in the struggle against violence").215. See id.216. See id. at 119-20 (suggesting that festivals "commemorate a sacrificial crisis"

in order to bring the community together).

526 [Vol. 38:487



that it would be consequentialist, but it would not assume thatthose inclined to murder make rational cost-benefit calculations.If the sacrificial ritual is effective, it will reduce the level of vio-lence in the community, but not by altering people's perceptionsof the cost of committing violence. 217 Rather, it will reduce thelevel of violence by breaking the cycle of imitative violence, andby creating a sense of unity in the community in its oppositionto the executed criminal.218 If this is so, the effect may be re-flected in indices far beyond the community's murder rate. Inter-estingly enough, one recent study purports to show that thecommunity's use of the death penalty, although it does not serveto reduce the homicide rate, is correlated with a reduction inlesser crimes.1 9 In terms of classic deterrence theory, thiswould seem to be a mere coincidence. How can increasing thepenalty for act A deter people from committing act B? If deter-rence is not a matter of rational cost-benefit analysis, however,but rather a consequence of reduction in the need for membersof the community to act in hostile, aggressive ways generally, itmay be more than coincidence.

In short, retentionists could argue that Girard has demon-strated that societal cohesion requires at least occasional formaluse of the ultimate penalty in order to unify the community. Ifthis argument is accepted as true, then it easily follows that theproper victim of such a sacrifice should be one who has set him-self deliberately against community norms in the most radicalway possible, by deliberately taking the life of another. To aban-don the death penalty, then, would be to deprive the communityof an important tool of social cohesion. Gregg v. Georgia2. rec-ognized that in a modern society committed to the rule of law,the victim must not be chosen in an arbitrary way, but rather in

217. See id. at 124 (proposing that ritual violence will not awaken hostility in the group).218. See id. (stating that [t]he community stands united before the onslaught of'evil spirits'.").219. See William C. Bailey & Ruth D. Peterson, Capital Punishment and Non.Capi-tal Crimes: A Test of Deterrence, General Prevention, and System-Overload Argu-ments, 54 ALB. L. REV. 681, 699 (1990) (stating that, on the basis of their research,they have determined that capital punishment has an effect on crimes such as rob-'bery, burglary, and assault when executions receive a certain amount of televisioncoverage).220. 428 U.S. 153 (1976).

5271997]



a way that is accepted by the community as rational."2 Thisreform was necessary in order to maintain the social acceptanceof capital punishment, without which it could not serve its unify-ing function. To go further, according to this retentionist argu-ment, would be dangerous because the abandonment of thedeath penalty could foster disunity.

Basically, a retentionist might accept Girard's work insofar asit is descriptive, but reject his call for a renunciation of all formsof sacrificial community violence as dangerous and utopian.Even the descriptive parts of Girard's theory will present prob-lems for retentionists, however, as will be seen in the followingsection.

2. Implications for Abolitionists

Abolitionist arguments have tended to take one of two differ-ent forms. The first, exemplified by Justices Brennan and Mar-shall, rejects capital punishment in theory as well as in prac-tice.22 The second, exemplified by the final death penalty opin-ions of Justice Blackmun,2" is more modest. Instead of at-tacking the fundamental theoretical underpinnings of capitalpunishment, it focuses on the empirical reality of the prac-tice. 4 The capriciousness and unfairness of that empiricalreality, and its failure to satisfy the demands of the theory es-tablish that even if the theory is sound, the practice is indefensi-

221. See id. at 188-95 (explaining that a jury will not be allowed to sentence adefendant in a "capricious or arbitrary" manner).222. See Furman v. Georgia, 408 U.S. 238, 286 (Brennan, J., concurring) (stating

that the imposition of death is a socially unacceptable denial of human dignity thatviolates the Eighth Amendment); id. at 342-70 (Marshall, J., concurring) (concludingthat the death penalty does not foster the goals of retribution, deterrence, preventingrecidivism, encouraging confessions, eugenics, or reducing state expenditures, andthat the public is morally opposed to capital punishment).223. Callins v. Collins, 510 U.S. 1141, 1145 (1994) (Blackmun, J., dissenting) ("It is

virtually self-evident to me now that no combination of procedural rules or substan-tive regulations ever can save the death penalty from its inherent constitutional defi-ciencies."); see also Sims v. California, 114 S. Ct. 2782 (1994) (Blackmun, J., dissent-ing to denial of cert.).224. "The problem is that the inevitability of factual, legal, and moral error givesus a system that we know must wrongly kill some defendants, a system that failsto deliver the fair, consistent, and reliable sentences of death required by the Con-stitution." Callins, 510 U.S. at 1145 (Blackmun, J., dissenting).

528 [Vol. 38:487



ble.' In the 1970s, centrist Justices struck down the preva-lent death penalty statutes, but continued to maintain thatdefensible procedures were possible." 6 Justice Blackmun, inessence, took this a step further, concluding that, as a practicalmatter, creating a death penalty that satisfies the demands ofits own underlying theory is impossible."7

On the surface, the Brennan-Marshall approach seems moreattractive to abolitionists. It is more "principled;" it refutesdeath penalty advocates more thoroughly. By disposing of theunderlying theory, it assures the end of capital punishment, notmerely its reform. The Blackmun approach is seen as concedingtoo much, as insufficiently grounded in principle; it is somewhatgrudgingly endorsed as a second-best approach. Perhaps, howev-er, the emphasis placed on these two lines of argument shouldbe reversed.

Gil Bailie calls attention to Girard's conclusion that the ritual-ized killing that unifies the community loses its effectiveness ifand when the community begins to question assumptions suchas the actual guilt of the victim.2" Girard states that the com-munity turns away from violence, then, not because it rejectsthe theory that one who has caused intolerable dissension by hisown acts of violence deserves to die, but rather because it seesthat its chosen victim is not really guilty of that at all.229 ForGirard, writing from a radical Christian perspective, the para-

225. See id. at 1155-57 (Blackmun, J., dissenting).226. See id. at 1147-57 (Blackmun, J., dissenting) (explaining the series of decisionsby the Supreme Court regarding capital punishment).227. See id. at 1157 (Blackmun, J., dissenting) (holding that capital punishment isunconstitutional).

From the moment when knowledge of the mechanism begins to spread,there can be no turning back. It is impossible to rehabilitate a sacrificialmechanism in the process of decomposition because growing awareness ofthese mechanisms is what decomposes them; any effort to interrupt orreverse the process can only be made at the cost of the knowledge beingdisseminated.

THINGS HIDDEN, supra note 142, at 128.228. Once we see a victim as a victim, we empathize, and "unanimity requires thatall empathy for the victim or victims of the violence be extinguished." BAILIE, supranote 142, at 16. Bailie comments on contemporary society's increasing tendency toidentify itself with victims, rather than with victors. See id. at 17-29.229. See THINGS HIDDEN, supra note 142, at 180-223.

1997] 529



digm here is the execution of Jesus." ° The community's execu-tion of one who is totally innocent leads those who can see themagnitude of the injustice to reject the very notion of returningviolence for violenceY Prior beliefs about the justice of retri-bution must give way to the harsh reality of its misuse.

As this Article has shown, the proposition that capital punish-ment advances the goal of deterrence significantly has been ex-tensively refuted. 2 Abolitionists, however, have had less suc-cess in refuting the theory of retribution. Perhaps it is best toturn away from attempts to establish that an undoubtedlyguilty, fully competent and volitional, deliberate murderer doesnot merit the death penalty and toward more persistent at-tempts to illustrate the gap between this theoretical propositionand the actual practice of capital punishment.

When Michigan, in 1847, became the first jurisdiction to aban-don capital punishment, it did so largely in response to publicrevulsion over the execution of a prisoner subsequently shown tobe innocent. 3 The most compelling arguments made againstthe death penalty in modern times often focus on its use in casesthat, if not involving the innocent,' at least involve killerswho seem less than fully competent or volitional.' Even whenthe death penalty is rejected in principle, the basis is likely to begrounded in religious conviction, rather than an attempt to ra-tionally refute retributive theory. 6

230. See id. at 202-20; BAILIE, supra note 142, at 217-33.231. See THINGS HIDDEN, supra note 142, at 202-20.232. See ZIMRING & HAWKINS, supra note 57, at 167-86 (concluding that any deter-

rent effect is slight at best).233. See Rachel Reynolds, Byline, Gannett News Service, Sept. 18, 1989, available

in LEXIS, News Library, Arcaws file (explaining that the death penalty was struckdown after an innocent man was hung for the rape and murder of a young woman).234. For an inventory of erroneously decided death penalty cases, see MICHAEL L.

RADELET ET AL., IN SPITE OF INNOCENCE 282-356 (1992).235. See, e.g., Penry v. Lynaugh, 492 U.S. 302 (1989); see generally EMILY F. REED,

THE PENRY PENALTY: CAPITAL PUNISHMENT AND OFFENDERS WITH MENTAL RETARDA-TION (1993) (criticizing the Supreme Court's decision allowing mentally retardedindividuals who have committed capital crimes to be sentenced to death).236. For example, see HELEN PREJEAN, DEAD MAN WALKING: AN EYEWITNESS AC-

COUNT OF THE DEATH PENALTY IN THE UNITED STATES (1993), providing a powerfulcritique of capital punishment that does not seek to deny the guilt of offenders, but,rather, is based explicitly upon the religious conviction that retaliation is wrong.

[Vol. 38:487530



Thus, it is unlikely that a direct assault on the theories putforward to support the death penalty will effectively refute it.Instead, a strategy that points out the empirical reality of thedeath penalty, and the ways in which it fails to satisfy the de-mands of the theories used to justify it, may effectively erode theability of executions to perform their function of unifying thecommunity. Girard, again writing from his radical Christian per-spective, sees the ultimate victory of those who reject retributiveviolence as flowing from the inevitable, if gradual and fitful,spread of the lessons learned in the aftermath of the executionof Christ. 7 The failure of the death penalty to satisfy rationaldemands placed on it by retributive theory, however, also can beargued in secular terms."s

Deterrence theory can be debated on the basis of empiricalevidence. 9 Even so, once the burden of proof is placed uponopponents of capital punishment, and set at a level that de-mands a great deal of certainty, the deterrence debate becomesfutile. It will never be possible to disprove deterrence with suffi-cient certainty to satisfy that burden.4 Further, as this Arti-cle has shown, deterrence theory seems less and less central tothe debate, even as it is understood currently. Most abolitionistsand retentionists seem committed to their positions regardless ofthe evidence on deterrence. Deterrence seems to be an argumentadded to bolster a position already taken.

On its face, retribution theory does not depend on empiricalevidence. Instead, retribution depends on the validity of theproposition that, as applied to capital punishment, one who de-liberately kills deserves to die. It may be futile to argue withthat central contention, as a matter of abstract theory. But theprocess of deciding who has killed, and more importantly, whohas killed with sufficient deliberation, is subject to empiricalanalysis. To demonstrate that the current system of applyingcapital punishment, or perhaps any practical system for doing

237. See THINGS HIDDEN, supra note 142, at 180-223.238. This approach is essentially the one taken by Justice Blackmun in his final

death penalty opinions. See supra notes 223-27 and accompanying text.239. For a discussion of empirical data and research methods used in developing

such data, see ZIMRING & HAWKINS, supra note 8, at 92-268.240. See Ud. at 327-36 (explaining the problems of measurement in deterrence studies).

1997] 531



so, will fall far short of its stated goal of identifying the mostculpable killers may effectively undermine faith in the sanction.Such a demonstration will not undermine the theoretical validi-ty of retributive notions, but instead will undermine the practi-cal ability to act as the theory demands. Justice Blackmun's ownchange of heart is the most prominent example of such a loss offaith.2" Furthermore, Girard's theories would predict thatwhen enough members of the community reach that conclusion,the death penalty will be unable to achieve its primary purposeof forging community unity.242

To conclude that the most effective way to attack the deathpenalty is to undermine confidence in its ability to satisfy inpractice the demands of retributive theory, however, leaves cru-cial questions unanswered. Girard's theories present abolition-ists with a significant challenge. If one accepts the notion thatritualized community violence plays an important role in bring-ing about and maintaining social cohesion,"'3 then the elimina-tion of the ritual, or the process of undermining its ability tounite the community, is not self-evidently a good thing. Girardwarns that a society deprived of its ritual violence may fallapart, because it will lack any method of putting an end to theintensifying cycle of imitative violence to which all societies aredrawn.2" Therefore, abolitionists must face the question ofwhether it is reasonable to risk these negative social conse-quences by undermining the death penalty as a means of achiev-ing social harmony.

Perhaps this question is somewhat academic. To a large ex-tent, the damage to the social function of the death penalty al-ready may have been done. Although majorities currently favorcapital punishment, the minority that opposes it is by no means

241. See supra notes 223-27 and accompanying text.242. See GIRARD, supra note 14, at 119-42.243. "[Dlemystification of the system necessarily coincides with the disintegration of

that system. .. . In fact, demystification leads to constantly increasing violence." Id.at 24. Unless the renunciation of present forms of violence itself repudiates violence,it may well be a step backward. See THINGS HIDDEN, supra note 142, at 245.244. The violence that replaces ritual violence may be "more energetic, more viru-lent, and the harbinger of something far worse-a violence that knows no bounds."GIRARD, supra note 14, at 24-25.

532 [Vol. 38:487



insignificant. 5 The disproportionate use of the death penaltyin cases involving racial minorities is accentuating division inthe overall community, rather than promoting unity.246 Out-spoken opposition to executions, either in general or in particu-lar, destroys the solemn unanimity of the community in its vio-lence toward the condemned individual, and in so doing, inter-feres with the ritual's effectiveness. To Girard, this oppositionresults from the inevitable spread of the Christian message.247

Whether it is a consequence of religious or secular conviction,and whether it is a trend that inevitably will grow or not, itseems clear that capital punishment does not have the power tobreak the cycle of social violence effectively.

When the ritual does not work, one response is to demandthat it be repeated more often, until it does work. Thus, it isunsurprising that the first few executions of the modern era ledto demands for more, and that the current fifty or so executionsthat occur each year have, in turn, merely led to demands formore, and swifter, instances of capital punishment.24 Never-theless, if the first few executions failed to unite the community,those that follow are also likely to fail. The question remainswhether the response of the community will be to turn awayfrom violence or to demand even more.

If one believes, as Girard does, in the ultimate and inevitabletriumph of the message of nonviolence,249 then perhaps one cango about the task of delegitimizing violent social institutions

245. One clear example of this minority viewpoint is expressed by Girard. See infranote 249 and accompanying text.246. See supra notes 78-79 and accompanying text.247. See GIRARD, supra note 14, at 119-42.248. See supra notes 80-81 and accompanying text (discussing the growing demand

for expediting the death penalty process).249.

[Elverything in the Kingdom of God comes down to the project of riddingmen of violence....

To leave violence behind, it is necessary to give up the idea of retribu-tion; it is therefore necessary to give up forms of conduct that have al-ways seemed to be natural and legitimate... . what must be given up isthe right to reprisals and even the right to what passes, in a number ofcases, for legitimate defence.

THINGS HIDDEN, supra note 142, at 197-98.

1997] 533



with no particular recommendation for how to replace them."soThis is especially so if the suggested replacement merely miti-gates, rather than eliminates, the use of violence as a tool ofsocial control. If one concedes, however, that the millennium isnot imminent,"' one must address in the near term the veryreal problem for abolitionists posed by Girard's theories. If theritual violence of capital punishment cannot unite the communi-ty, then what substitute can achieve that end?

At the very least, opposition to the death penalty should notbe presented merely as part of a broad denial of the legitimacyof social values widely held by the community, specifically thestrong disapproval of violent crime. Not only would such an ar-gument be unlikely to succeed, but if it were to succeed, it wouldin all likelihood leave a significant part of the community withenough unrelieved anger to make social unrest and backlashhighly probable. Thus, through both rhetoric and practical pro-posals, such as advocacy of life sentences without parole, a re-sponsible abolitionist argument must give the community a way,short of actual execution, to unify in its condemnation of violentcrimes. Nevertheless, although the need may be clear for suc-cessful alternatives to the ritual functions currently achievedthrough the death penalty, the specific question of the nature ofsuch alternatives needs much more thought.

In this regard, it is interesting to note that most western de-mocracies, for all practical purposes, have abolished the deathpenalty. 2 Yet, several of these countries retain it in theory for

250. Thus, Girard stated: "I am fully in favour of the major liquidation of philoso-phy and the sciences of man that is currently taking plae .... the real accomplish-ments of modem thought .. . are critical and negative." Id. at 135-36. Further, this"desacrilization" process must be followed by '[tihe definitive renunciation of violence,without any second thoughts . . . the condition sine qua non for the survival of hu-manity itself.. . ." Id. at 137.251. The millennium, in Christian theology, is the thousand-year period of peaceforeseen in the Book of Revelations. See JOHN F. HARRISON, THE SECOND COMING:POPULAR MILENARIANISM 1780-1850, at 3-10 (1979). Premillennialists believe thatthis period will follow the Second Coming of Christ. See id. at 4. Postmillennialists,in contrast, believe that the climax of the thousand-year period of righteousness willbe the Second Coming of Christ. See id. Contrary to widespread belief, themillenium does not have any necessary connection with the expiration of any partic-ular thousand-year period calculated from the birth of Christ. See id. at 5, 7.252. See ZIMBING & HAWKINS, supra note 57, at 4-6.

534 [Vol. 38:487



some exceptionally rare instances, such as certain acts commit-ted in time of -war.' Perhaps the mere presence of the ulti-mate sanction on the statute books, even if it is never, or hardlyever invoked, helps to satisfy the need for social solidarity. Itdoes, after all, proclaim the community's belief that someone,even if that person is only hypothetical, justifiably is the targetof the whole community's wrath. Thus, the fact that the deathpenalty is available in Israel for those guilty of Nazi atrocitiesmay well serve a salutary social purpose, even though only oneman has ever been executed under this provision.'

Perhaps, then, the practical abolition of the death penalty willrequire a concession that it is still available in theory. This re-turns the discussion, in a way, to what for many is the begin-ning of the death penalty debate: the Bible. As previously noted,the Bible solemnly prescribes death as appropriate punishment,but, in practice, ancient Israel constructed procedures to assureinfrequent application. 5 Thus, the community satisfied itsneed to afirm its solidarity in opposing the crime without re-quiring actual, frequent executions. If one embraces this theory,current trends in the United States are disturbing. Increasingly,calls are being made, not only for the enactment of death pen-alty statutes, but for streamlined procedures that will assure thecompletion of more executions. 6

Unfortunately, then, the question of whether effective alterna-tive rituals can serve the same social function of uniting thecommunity in opposition to a common enemy must remain open.Girard's theories suggest that the question of alternative ritualsmust be addressed if real progress is to be made in furtheringthe abolitionist cause.

V. CONCLUSION

In a legal environment that assumes rational argument to bethe foundation of governmental decision making, it is easy tooverlook the possibility that elements of the legal system are

253. See id. at 5.254. See id. at 6. The exception, of course, was Nazi war criminal AdolfEichmann. See id.255. See supra notes 122-24 and accompanying text.256. See supra notes 127-32 and accompanying text.

1997] 535



best explained by powerful, nonrational motives. The contempo-rary debate over capital punishment proceeds, as it has for cen-turies, with a focus on rational justifications for the practice.Yet, powerful evidence that the practice does not advance therational goals set forth as its objectives should make one wonderabout society's basic assumptions concerning the death penalty'sreal function.

Although commentators have noted the symbolic function ofexecutions, this usually has been relegated to the margins of thedebate; unsurprisingly, the debate has focused on the purportedgoals of deterrence and retribution. Upon close examination,however, these goals seem strangely beside the point. Support-ers of the death penalty are unmoved by evidence that it doesnot deter violent crime, or by evidence that in practice the deathpenalty falls woefully short of the demands of retributive theory.

Perhaps, then, the key to understanding the death penalty isto see it as filling a nonrational, yet very real, need. Ren6Girard's theory of mimetic violence provides an intriguing alter-native explanation for the death penalty, and one that also ex-plains its resistance to rational argument. If Girard is correct,the death penalty serves primarily as a ritual of violencethrough which the community attempts to unify itself, concen-trating its urge to mimetic violence on one individual identifiedas the other, the enemy of the community. If the death penaltyshould be understood in this way then one must view the sur-rounding debate in a fundamentally different way.

The debate over the deterrent effect of capital punishmentseems to be largely irrelevant. Not only has the Supreme Courtdecided that a showing of deterrence is unnecessary, 1 7 butmost death penalty supporters, even if they believe in the deter-rent effect of the death penalty, do not seem to rest their sup-port of capital punishment on the existence of such an effect. Asretribution becomes more central to the justification of capitalpunishment, it is unsurprising that abolitionists find themselvesfeeling the need to challenge the fundamental notion that a de-liberate, fully volitional killer deserves to die under any circum-stances.

257. See Edmund v. Florida, 458 U.S. 782, 798-99 (1982).

536 [Vol. 38:487



If Girard's theories explain capital punishment, then the truthor falsity of the retribution proposition is not really at the heartof the practice. The death penalty is, at its core,consequentialist, but not in the sense that it will deter potentialkillers by getting them to adjust their mental cost-benefit calcu-lations. Instead, it seeks to reduce the amount of violence in thecommunity by creating communal unity in opposition to the exe-cuted killer. The ability of the death penalty to achieve thoseconsequentialist goals is largely dependent upon the extent towhich the community believes that it is applied consistentlywith deeply held retributivist ideas.

Rather than launching a frontal assault on the theoreticallegitimacy of the death penalty, particularly under retributivetheory, perhaps the most effective way to attack the death pen-alty is to expose its failings, in the real world, to satisfy the de-mands of retributive theory. Does the present system actuallydetermine, in a reliable way, the subset of killers who are mostculpable? Can any system do so?" 8 If these questions continueto be raised effectively, they may seriously interfere with theability of the death penalty to serve its central purpose of unit-ing the community. Of course, this route holds its own dangers.One response to the failure of the death penalty to "work" mightbe not to abandon it, but rather to expand it, in an attempt tosomehow get it to work. After all, its proponents are unclearabout just why it once worked, and so they will fail to under-stand the futility of trying to make a divisive death penalty dothe work of unifying the community. Furthermore, the need forthe unifying ritual to break the cycle of mimetic violence willremain.

Retentionists, of course, can draw their own conclusions fromGirard's theories. They may argue that his theories merely dem-onstrate the important function that capital punishment serves,

258. Jewish law, which set up such high procedural barriers to executions as tomake them nearly impossible, did so in light of the fallibility of any human judg-ment, however careful. "In effect, the law demands a differentiation between thequality of God's omniscient justice and the fragile, fallible justice of the worldlycourts." Rosenberg & Rosenberg, supra note 122, at 620. Thus, the inevitability ofimperfection in judging guilt weighed heavily against imposing punishment, evenwhen guilt seemed probable. See id.

1997] 537


538 WILLIAM AND MARY LAW REVIW [Vol. 38:487

one that is abandoned only at the risk of removing an importanttool to further peace in the community. Rituals are not merelyamusing sideshows; they are crucial to any community.

Opponents of the death penalty must accept the challenge ofproviding alternative ways of allowing the community to effec-tively express its unity in opposing violent crime. Finding suchalternatives will not be easy. In a diverse, contentious, frag-mented nation, forging unifying rituals that are effective will bea challenge. The first step is to realize that such a task is neces-sary. That realization will require serious consideration of thepossibility that the rationalist arguments over deterrence andretribution, which both sides assume are at the core of the deathpenalty debate, may be far less important than they seem.


Documents

What's Guilt (or Deterrence) Got to Do With It?: The Death